You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2023/04/21 18:27:19 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@af17cfeb4c71fdfb7f062238859abdf1479fa48f)
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 f35cf23b3e deploying docs (apache/tvm@af17cfeb4c71fdfb7f062238859abdf1479fa48f)
f35cf23b3e is described below
commit f35cf23b3e303b2a34b5fb293b25818bb7d29b8d
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Fri Apr 21 18:27:12 2023 +0000
deploying docs (apache/tvm@af17cfeb4c71fdfb7f062238859abdf1479fa48f)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 349354 -> 347925 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 24292 -> 24009 bytes
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 2 +-
.../how_to/compile_models/from_paddle.rst.txt | 5 +
.../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_model_on_adreno_tvmc.rst.txt | 2 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 22 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 10 +-
.../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 | 1759 ++++++++++----------
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 102 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 12 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 682 ++++----
.../work_with_microtvm/micro_autotune.rst.txt | 18 +-
.../work_with_microtvm/micro_pytorch.rst.txt | 4 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 18 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 14 +-
.../work_with_relay/sg_execution_times.rst.txt | 10 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 4 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 11 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 62 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 22 +-
.../tutorial/tensor_expr_get_started.rst.txt | 44 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_darknet.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 15 +-
docs/how_to/compile_models/from_paddle.html | 1 +
docs/how_to/compile_models/from_pytorch.html | 13 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_adreno.html | 2 +-
.../deploy_models/deploy_model_on_adreno_tvmc.html | 16 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 55 +-
docs/how_to/deploy_models/deploy_prequantized.html | 7 +-
.../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 | 37 +-
docs/how_to/deploy_models/sg_execution_times.html | 22 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 10 +-
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 | 1759 ++++++++++----------
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 102 +-
.../tune_with_autotvm/sg_execution_times.html | 12 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 682 ++++----
docs/how_to/work_with_microtvm/micro_autotune.html | 18 +-
docs/how_to/work_with_microtvm/micro_pytorch.html | 5 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 14 +-
.../how_to/work_with_relay/sg_execution_times.html | 10 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
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 +-
docs/reference/api/typedoc/classes/instance.html | 58 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
.../api/typedoc/classes/runtimecontext.html | 22 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
docs/reference/api/typedoc/classes/tvmarray.html | 16 +-
docs/reference/api/typedoc/classes/tvmobject.html | 12 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 124 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 4 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 7 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 276 +--
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 26 +-
docs/tutorial/tensor_expr_get_started.html | 44 +-
131 files changed, 3399 insertions(+), 3443 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index b147d290fc..afcc2f6f8b 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 a474f132c9..027a11bd69 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 97d858c19a..492dbcc3fc 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 25.899 seconds)
+ **Total running time of the script:** ( 1 minutes 27.632 seconds)
.. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index b03d67a7aa..be2780855f 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.zipc1772097-6190-47d0-863c-9dddc2a5c20e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip8b605d40-69c4-4d1a-b2c5-692bee6cd627 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 5832f06abc..ee0de7471f 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, 63.1MB/s]
30%|##9 | 12.3M/41.5M [00:00<00:00, 47.0MB/s]
41%|####1 | 17.0M/41.5M [00:00<00:00, 45.0MB/s]
54%|#####3 | 22.3M/41.5M [00:00<00:00, 47.4MB/s]
65%|######4 | 26.9M/41.5M [00:00<00:00, 41.9MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 43.3MB/s]
92%|#########2| 38.3M/41.5M [00:01<00:00, 32.7MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 37.2MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
15%|#5 | 6.33M/41.5M [00:00<00:00, 38.5MB/s]
24%|##4 | 10.0M/41.5M [00:00<00:01, 31.0MB/s]
39%|###8 | 16.0M/41.5M [00:00<00:00, 34.8MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 45.1MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 48.1MB/s]
92%|#########2| 38.3M/41.5M [00:00<00:00, 43.8MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 42.9MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_paddle.rst.txt b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
index 8df75c9e33..b886f5172e 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -207,6 +207,11 @@ Look up prediction top 1 index in 1000 class synset.
+.. rst-class:: sphx-glr-timing
+
+ **Total running time of the script:** ( 1 minutes 2.603 seconds)
+
+
.. _sphx_glr_download_how_to_compile_models_from_paddle.py:
.. only:: html
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 50f5267c08..b143a0c3fb 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, 70.2MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 55.9MB/s]
54%|#####3 | 24.0M/44.7M [00:00<00:00, 61.2MB/s]
72%|#######1 | 32.0M/44.7M [00:00<00:00, 60.3MB/s]
86%|########5 | 38.3M/44.7M [00:00<00:00, 35.5MB/s]
96%|#########5| 42.7M/44.7M [00:01<00:00, 32.4MB/s]
100%|##########| 44.7M/44.7M [00:01<00:00, 42.0MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 53.0MB/s]
32%|###2 | 14.3M/44.7M [00:00<00:00, 55.1MB/s]
49%|####9 | 22.1M/44.7M [00:00<00:00, 65.4MB/s]
69%|######9 | 30.9M/44.7M [00:00<00:00, 75.1MB/s]
86%|########5 | 38.3M/44.7M [00:00<00:00, 72.1MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 63.2MB/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 14e6200281..e22affed8e 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -430,7 +430,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 30.395 seconds)
+ **Total running time of the script:** ( 1 minutes 36.993 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 5284efc001..71a1700e45 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
=================
-**07:01.986** total execution time for **how_to_compile_models** files:
+**07:22.248** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:30.395 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:36.993 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:25.899 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:27.632 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:58.807 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 01:02.603 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:38.578 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:40.561 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:33.808 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:35.433 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:31.132 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:32.282 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:28.431 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:29.077 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:26.973 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:28.335 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:25.132 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:26.464 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.831 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.868 | 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 efd74e2798..0c839a719b 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
@@ -673,7 +673,7 @@ 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)
- 2752.7610 2752.9445 2755.0147 2751.1536 1.2200
+ 2754.4285 2753.8857 2758.8902 2751.5141 2.0346
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
index b11fb682c4..6b94438898 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
@@ -127,7 +127,7 @@ Make a Keras Resnet50 Model
.. code-block:: none
Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
-
8192/102967424 [..............................] - ETA: 0s
8380416/102967424 [=>............................] - ETA: 1s
15024128/102967424 [===>..........................] - ETA: 0s
16769024/102967424 [===>..........................] - ETA: 1s
19611648/102967424 [====>.........................] - ETA: 1s
22970368/102967424 [=====>........................] - ETA: 1s
23412736/102967424 [=====>........................] - ETA: 1s
25157632/102967424 [======>.......................] - ETA: 1s
31498240/102967424 [========>.....................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
40189952/102967424 [==========>...................] - ETA: 1s
41934848/102967424 [===========>..................] - ETA: 1s
48578560/102967424 [=============>................] - ETA: 1s
50323456/102967424 [=============>................] - ETA: 1s
58712064/102967424 [================>.............] - ETA: 1s
65355776/102967424 [==================>...........] -
ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] - ETA: 0s
75489280/102967424 [====================>.........] - ETA: 0s
82124800/102967424 [======================>.......] - ETA: 0s
83877888/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
98910208/102967424
[===========================>..] - ETA: 0s
102850560/102967424 [============================>.] - ETA: 0s
102967424/102967424 [==============================] - 2s 0us/step
+
8192/102967424 [..............................] - ETA: 0s
8380416/102967424 [=>............................] - ETA: 1s
15024128/102967424 [===>..........................] - ETA: 1s
16769024/102967424 [===>..........................] - ETA: 1s
24027136/102967424 [======>.......................] - ETA: 1s
25157632/102967424 [======>.......................] - ETA: 1s
31899648/102967424 [========>.....................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
41934848/102967424 [===========>..................] - ETA: 1s
50135040/102967424 [=============>................] - ETA: 0s
50323456/102967424 [=============>................] - ETA: 1s
56967168/102967424 [===============>..............] - ETA: 0s
58712064/102967424 [================>.............] - ETA: 0s
65355776/102967424 [==================>...........] - ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] -
ETA: 0s
75489280/102967424 [====================>.........] - ETA: 0s
82124800/102967424 [======================>.......] - ETA: 0s
83877888/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
100646912/102967424 [============================>.] - ETA: 0s
102850560/102967424 [============================>.] - ETA: 0s
102967424/102967424
[==============================] - 2s 0us/step
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index f701b0a14c..c1d142fce4 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)
- 16.3212 16.2892 16.4534 16.2165 0.0857
+ 16.8435 16.8438 17.2434 16.6309 0.1515
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 16814cc0a7..d39da0099b 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]
4%|3 | 6.30M/170M [00:00<00:03, 44.0MB/s]
6%|6 | 10.5M/170M [00:00<00:03, 43.1MB/s]
9%|8 | 14.6M/170M [00:00<00:05, 29.8MB/s]
10%|# | 17.7M/170M [00:00<00:05, 29.1MB/s]
14%|#4 | 24.0M/170M [00:00<00:04, 33.8MB/s]
19%|#8 | 32.0M/170M [00:00<00:03, 39.2MB/s]
24%|##3 | 40.0M/170M [00:01<00:03, 42.8MB/s]
28%|##8 | 48.0M/170M [00:01<00:02, 43.5MB/s]
31%|### | 52.1M/170M [00:01<00:03, 36.7MB/s]
33%|###2 | 56.0M/170M [00:01<00:03, 36.7MB/s]
38%|###7 | 64.0M/170M [00:01<00:03, 36.6MB/s]
42%|####2 | 72.0M/170M [00:01<00:02, 42.4MB/s]
47%|####7 | 80.0M/170M [00:02<00:02, 46.9MB/s]
52%|#####1 | 88.0M/170M [00:02<00:01, 48.7MB/s]
56%|#####5 | 94.3M/170M [00:02<00:01, 52.2MB/s]
59%|#####8 | 99.5M/170M [00:02<00:01, 52.3MB/s]
62%|######1 | 105M/170M [00:02<00:01, 45.6MB/s
]
66%|######5 | 112M/170M [00:02<00:01, 49.8MB/s]
71%|####### | 120M/170M [00:02<00:00, 57.4MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 56.4MB/s]
80%|######## | 136M/170M [00:03<00:00, 52.9MB/s]
85%|########4 | 144M/170M [00:03<00:00, 51.0MB/s]
89%|########9 | 152M/170M [00:03<00:00, 45.0MB/s]
94%|#########4| 160M/170M [00:03<00:00, 46.1MB/s]
98%|#########7| 166M/170M [00:03<00:00, 48.4MB/s]
100%|##########| 170M/170M [00:03<00:00, 45.1MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:02, 59.8MB/s]
8%|8 | 14.3M/170M [00:00<00:03, 48.0MB/s]
11%|#1 | 19.0M/170M [00:00<00:03, 43.0MB/s]
14%|#3 | 23.1M/170M [00:00<00:04, 38.4MB/s]
16%|#5 | 26.7M/170M [00:00<00:04, 36.7MB/s]
19%|#8 | 32.0M/170M [00:00<00:04, 33.2MB/s]
24%|##3 | 40.0M/170M [00:01<00:03, 42.3MB/s]
27%|##7 | 46.3M/170M [00:01<00:02, 46.8MB/s]
30%|### | 51.0M/170M [00:01<00:03, 39.2MB/s]
33%|###2 | 56.0M/170M [00:01<00:03, 38.5MB/s]
37%|###6 | 62.3M/170M [00:01<00:02, 43.6MB/s]
40%|#### | 68.7M/170M [00:01<00:02, 49.2MB/s]
43%|####3 | 73.7M/170M [00:01<00:02, 46.4MB/s]
47%|####7 | 80.0M/170M [00:01<00:02, 41.6MB/s]
52%|#####1 | 88.0M/170M [00:02<00:02, 40.1MB/s]
56%|#####5 | 94.3M/170M [00:02<00:01, 45.4MB/s]
60%|#####9 | 101M/170M [00:02<00:01, 51.8MB/s
]
63%|######2 | 107M/170M [00:02<00:01, 36.5MB/s]
66%|######5 | 112M/170M [00:02<00:01, 38.2MB/s]
71%|####### | 120M/170M [00:02<00:01, 46.6MB/s]
75%|#######5 | 128M/170M [00:03<00:00, 47.3MB/s]
80%|######## | 136M/170M [00:03<00:00, 49.6MB/s]
87%|########6 | 148M/170M [00:03<00:00, 65.5MB/s]
91%|#########1| 155M/170M [00:03<00:00, 54.8MB/s]
95%|#########4| 161M/170M [00:03<00:00, 50.5MB/s]
98%|#########7| 166M/170M [00:03<00:00, 51.5MB/s]
100%|##########| 170M/170M [00:03<00:00, 45.8MB/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 49.647 seconds)
+ **Total running time of the script:** ( 4 minutes 4.015 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 e5604e9f35..7c07d10635 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, 39.5MB/s]
90%|########9 | 12.2M/13.6M [00:00<00:00, 28.5MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 33.1MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 39.5MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 46.5MB/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)
- 90.2833 90.2104 91.8116 90.0660 0.2546
+ 90.7178 90.5539 103.0548 90.2578 1.2598
@@ -458,7 +458,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 20.763 seconds)
+ **Total running time of the script:** ( 1 minutes 23.693 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 36bd41d7f1..d53b4197e5 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)
- 119.9694 119.9506 121.3017 119.1650 0.3862
+ 122.4928 122.4673 125.0076 121.6657 0.3962
@@ -460,7 +460,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 31.029 seconds)
+ **Total running time of the script:** ( 2 minutes 37.759 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 f1bec2e780..65618000a1 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 53.156 seconds)
+ **Total running time of the script:** ( 1 minutes 49.957 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 bc3337893c..03419ed062 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 | 5596/132723 [00:00<00:02, 55955.38KB/s]
10%|9 | 13038/132723 [00:00<00:01, 66812.84KB/s]
15%|#5 | 20544/132723 [00:00<00:01, 70575.90KB/s]
21%|##1 | 28379/132723 [00:00<00:01, 73642.54KB/s]
27%|##7 | 36257/132723 [00:00<00:01, 75493.91KB/s]
33%|###3 | 44136/132723 [00:00<00:01, 76612.75KB/s]
39%|###9 | 52063/132723 [00:00<00:01, 77479.88KB/s]
45%|####5 | 59974/132723 [00:00<00:00, 77995.99KB/s]
51%|#####1 | 67774/132723 [00:01<00:01, 58530.94KB/s]
57%|#####6 | 75637/132723 [00:01<00:00, 63573.63KB/s]
62%|######2 | 82588/132723 [00:01<00:00, 55777.49KB/s]
68%|######7 | 89993/132723 [00:01<00:00, 60235.43KB/s]
73%|#######3 | 97548/132723 [00:01<00:00, 64192.59KB/s]
79%|#######8 | 104377/132723 [00:01<00:00, 50779.77KB/s]
84%|########4 | 111954/132723 [00:01<00:00, 56540.28KB/s]
89%|########9
| 118267/132723 [00:01<00:00, 52465.75KB/s]
95%|#########4| 125839/132723 [00:02<00:00, 58091.76KB/s]
100%|##########| 132723/132723 [00:02<00:00, 62888.07KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
4%|4 | 5789/132723 [00:00<00:02, 57884.26KB/s]
11%|# | 14260/132723 [00:00<00:01, 73657.72KB/s]
17%|#7 | 22751/132723 [00:00<00:01, 78793.31KB/s]
24%|##3 | 31310/132723 [00:00<00:01, 81472.92KB/s]
30%|##9 | 39780/132723 [00:00<00:01, 82632.96KB/s]
36%|###6 | 48325/132723 [00:00<00:01, 83585.43KB/s]
43%|####2 | 56684/132723 [00:00<00:00, 83499.49KB/s]
49%|####9 | 65134/132723 [00:00<00:00, 83814.16KB/s]
55%|#####5 | 73516/132723 [00:00<00:00, 74425.53KB/s]
62%|######1 | 81905/132723 [00:01<00:00, 74779.40KB/s]
68%|######8 | 90329/132723 [00:01<00:00, 77444.72KB/s]
74%|#######4 | 98288/132723 [00:01<00:00, 66600.20KB/s]
79%|#######9 | 105269/132723 [00:01<00:00, 65115.59KB/s]
86%|########5 | 113722/132723 [00:01<00:00, 70229.24KB/s]
91%|#########1| 120969/132723 [00:01<00:00, 68298.71KB/s]
97%|########
#7| 129306/132723 [00:01<00:00, 72419.53KB/s]
100%|##########| 132723/132723 [00:01<00:00, 74499.20KB/s]
@@ -246,7 +246,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 8.954 seconds)
+ **Total running time of the script:** ( 4 minutes 17.961 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 a016af8227..1376534787 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,28 +5,28 @@
Computation times
=================
-**17:21.013** total execution time for **how_to_deploy_models** files:
+**17:58.983** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 04:08.954 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 04:17.961 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:49.647 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 04:04.015 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:31.029 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:37.759 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:53.156 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:49.957 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:20.763 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:23.693 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:58.545 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:59.738 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:54.934 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:57.007 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:45.199 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:47.956 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:29.648 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:30.831 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:29.133 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:30.060 | 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 27a6107d63..049895bef9 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.zip24d80f42-0c33-45cf-8eaf-9515a068bd6e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip37ec2eff-08b4-4e2b-b5c9-0b8bfaa790a7 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 4e1e124ca4..2ed3602d42 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
=================
-**00:57.180** total execution time for **how_to_extend_tvm** files:
+**00:59.464** 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``) | 00:53.201 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:55.300 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.856 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.983 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.116 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.173 | 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 |
+| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.008 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index 3d3a0df584..abb805a1f4 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: 23154us [23154us] (48.55%; 48.55%)
- FoldScaleAxis: 24538us [9us] (51.45%; 51.45%)
- FoldConstant: 24529us [1769us] (51.43%; 99.96%)
- InferType: 22760us [22760us] (47.72%; 92.79%)
+ InferType: 23568us [23568us] (48.55%; 48.55%)
+ FoldScaleAxis: 24980us [9us] (51.45%; 51.45%)
+ FoldConstant: 24971us [1838us] (51.44%; 99.96%)
+ InferType: 23133us [23133us] (47.65%; 92.64%)
@@ -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: 22816us [22816us] (48.50%; 48.50%)
- FoldScaleAxis: 24224us [8us] (51.50%; 51.50%)
- FoldConstant: 24216us [1799us] (51.48%; 99.97%)
- InferType: 22417us [22417us] (47.66%; 92.57%)
+ InferType: 23479us [23479us] (48.54%; 48.54%)
+ FoldScaleAxis: 24896us [8us] (51.46%; 51.46%)
+ FoldConstant: 24888us [1833us] (51.45%; 99.97%)
+ InferType: 23055us [23055us] (47.66%; 92.64%)
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 7db43819f3..edfe03fc77 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -331,7 +331,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 53.512065 ms
+ Convolution: 53.535198 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 bb563bdefd..794fe7463d 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -598,7 +598,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 12.237981 ms
+ conv2d with tensor core: 12.243171 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 8fed73d70b..50de75e680 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: 0.018904
- Baseline: 3.395954
+ Numpy running time: 0.020891
+ Baseline: 3.647581
@@ -227,7 +227,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.318500
+ Opt1: 0.347496
@@ -318,7 +318,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.359258
+ Opt2: 0.369314
@@ -406,7 +406,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.120890
+ Opt3: 0.141369
@@ -523,7 +523,7 @@ flattening.
.. code-block:: none
- Opt4: 0.110020
+ Opt4: 0.113683
@@ -635,7 +635,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111328
+ Opt5: 0.115145
@@ -748,7 +748,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.147865
+ Opt6: 0.150701
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 61bbf4d4a4..8e31eee2c2 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
=================
-**00:36.023** total execution time for **how_to_optimize_operators** files:
+**00:38.176** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:33.011 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:34.960 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.879 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:02.015 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.133 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.201 | 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 bcce24b51f..a623bb8932 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,18 +5,18 @@
Computation times
=================
-**10:22.795** total execution time for **how_to_tune_with_autoscheduler** files:
+**10:56.303** total execution time for **how_to_tune_with_autoscheduler** files:
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:19.519 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:45.477 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:46.122 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:49.529 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:13.644 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:16.077 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:34.304 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:34.832 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:14.874 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:15.551 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:14.333 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:14.837 | 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 3a0791f34e..26a1913723 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -243,12 +243,12 @@ cooperative fetching, unrolling and operator fusion.
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- blockIdx_x = T.launch_thread("blockIdx.x", 28)
- conv2d_nchw = T.allocate([14], "float32", "local")
- pad_temp_shared = T.allocate([72], "float32", "shared")
- kernel_shared = T.allocate([3072], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 64)
- conv2d_nchw_1 = T.Buffer((14,), data=conv2d_nchw, scope="local", align=32)
+ blockIdx_x = T.launch_thread("blockIdx.x", 112)
+ conv2d_nchw = T.allocate([7], "float32", "local")
+ pad_temp_shared = T.allocate([576], "float32", "shared")
+ kernel_shared = T.allocate([6144], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 32)
+ conv2d_nchw_1 = T.Buffer((7,), data=conv2d_nchw, scope="local", align=16)
conv2d_nchw_1[0] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
@@ -256,466 +256,536 @@ cooperative fetching, unrolling and operator fusion.
conv2d_nchw_1[4] = T.float32(0)
conv2d_nchw_1[5] = T.float32(0)
conv2d_nchw_1[6] = T.float32(0)
- conv2d_nchw_1[7] = T.float32(0)
- conv2d_nchw_1[8] = T.float32(0)
- conv2d_nchw_1[9] = T.float32(0)
- conv2d_nchw_1[10] = T.float32(0)
- conv2d_nchw_1[11] = T.float32(0)
- conv2d_nchw_1[12] = T.float32(0)
- conv2d_nchw_1[13] = T.float32(0)
- for rc_outer_outer, ry_outer_outer in T.grid(64, 3):
- cse_var_2: T.int32 = rc_outer_outer * 72
- cse_var_1: T.int32 = ry_outer_outer * 3
- pad_temp_shared_1 = T.Buffer((72,), data=pad_temp_shared, scope="shared")
- with T.launch_thread("threadIdx.x", 64) as threadIdx_x_1:
- data_1 = T.Buffer((25088,), data=data.data)
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[rc_outer_outer * 392 + threadIdx_x_1 * 4 // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + threadIdx_x_1 * 4 % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 1] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 1) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 2] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 2) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 3] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 3) % 9 - 8], T.float32(0))
+ for rc_outer_outer, rx_outer_outer in T.grid(8, 3):
+ cse_var_1: T.int32 = rc_outer_outer * 3136
threadIdx_x_1 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((576,), data=pad_temp_shared, scope="shared")
+ data_1 = T.Buffer((25088,), data=data.data)
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + threadIdx_x_1 // 9 * 49 + threadIdx_x_1 % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 32] = T.if_then_else(1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 32) // 9 * 49 + (threadIdx_x_1 + 5) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 64] = T.if_then_else(1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 64) // 9 * 49 + (threadIdx_x_1 + 1) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 96] = T.if_then_else(1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 96) // 9 * 49 + (threadIdx_x_1 + 6) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 128] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 128) // 9 * 49 + (threadIdx_x_1 + 2) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 160] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 160) // 9 * 49 + (threadIdx_x_1 + 7) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 192] = T.if_then_else(1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 192) // 9 * 49 + (threadIdx_x_1 + 3) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 224) // 9 * 49 + (threadIdx_x_1 + 8) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 256] = T.if_then_else(1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 256) // 9 * 49 + (threadIdx_x_1 + 4) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 288] = T.if_then_else(1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + threadIdx_x_1 // 9 * 49 + threadIdx_x_1 % 9 * 7 + rx_outer_outer + blockIdx_x % 7 + 1560], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 320] = T.if_then_else(1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 320) // 9 * 49 + (threadIdx_x_1 + 5) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 352] = T.if_then_else(1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 352) // 9 * 49 + (threadIdx_x_1 + 1) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 384] = T.if_then_else(1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 384) // 9 * 49 + (threadIdx_x_1 + 6) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 416] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 416) // 9 * 49 + (threadIdx_x_1 + 2) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 448) // 9 * 49 + (threadIdx_x_1 + 7) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 480] = T.if_then_else(1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 480) // 9 * 49 + (threadIdx_x_1 + 3) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 512] = T.if_then_else(1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 512) // 9 * 49 + (threadIdx_x_1 + 8) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 544] = T.if_then_else(1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 544) // 9 * 49 + (threadIdx_x_1 + 4) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ threadIdx_x_2 = T.env_thread("threadIdx.x")
+ kernel_shared_1 = T.Buffer((6144,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 64) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 64) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 128) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 128) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 192] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 36864]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 256) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 256) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 320) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 320) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 384] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 73728]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 448) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 448) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 512) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 512) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 576] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 110592]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 640) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 640) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 704) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 704) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 768] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 147456]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 832) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 832) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 896) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 896) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 960] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 184320]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1024) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1024) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1088) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1088) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1152] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 221184]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1216) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1216) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1280) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1280) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1344] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 258048]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1408) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1408) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1472) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1472) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1536] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 294912]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1600) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1600) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1664) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1664) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1728] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 331776]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1792) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1792) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1856) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1856) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1920] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 368640]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1984) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1984) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2048) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2048) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2112] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 405504]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2176) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2176) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2240) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2240) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2304] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 442368]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2368) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2368) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2432) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2432) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2496] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 479232]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2560) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2560) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2624) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2624) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2688] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 516096]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2752) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2752) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2816) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2816) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2880] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 552960]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2944) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2944) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 3008) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 3008) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 47]
- for i1_inner, i3_inner in T.grid(2, 7):
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 32] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 64] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 192]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 96] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 288]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 128] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 384]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 160] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 480]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 192] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4608]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 224] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4704]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 256] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4800]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 288] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4896]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 320] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4992]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 352] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 5088]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 384] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9216]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 416] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9312]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 448] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9408]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 480] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9504]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 512] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9600]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 544] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9696]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 576] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 13824]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 608] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 13920]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 640] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14016]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 672] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14112]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 704] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14208]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 736] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14304]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 768] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18432]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 800] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18528]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 832] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18624]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 864] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18720]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 896] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18816]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 928] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18912]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 960] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23040]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 992] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23136]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1024] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23232]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1056] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23328]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1088] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23424]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1120] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23520]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1152] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27648]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1184] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27744]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1216] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27840]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1248] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27936]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1280] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 28032]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1312] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 28128]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32256]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1376] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32352]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1408] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32448]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1440] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32544]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1472] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32640]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1504] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32736]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1536] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 36864]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 36960]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1600] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37056]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1632] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37152]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1664] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37248]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1696] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37344]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1728] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41472]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1760] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41568]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1792] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41664]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1824] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41760]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1856] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41856]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1888] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41952]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1920] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46080]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1952] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46176]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1984] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46272]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2016] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46368]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2048] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46464]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2080] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46560]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2112] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50688]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2144] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50784]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2176] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50880]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2208] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50976]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2240] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 51072]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2272] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 51168]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2304] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55296]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2336] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55392]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2368] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55488]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2400] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55584]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2432] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55680]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2464] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55776]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2496] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 59904]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2528] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60000]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2560] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60096]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2592] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60192]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2624] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60288]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2656] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60384]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64512]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2720] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64608]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2752] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64704]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2784] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64800]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2816] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64896]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2848] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64992]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2880] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69120]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2912] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69216]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2944] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69312]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2976] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69408]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3008] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69504]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3040] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69600]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3072] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 73728]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3104] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 73824]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3136] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 73920]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3168] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 74016]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3200] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 74112]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3232] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 74208]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3264] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78336]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3296] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78432]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3328] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78528]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3360] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78624]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3392] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78720]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3424] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78816]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3456] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 82944]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3488] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83040]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3520] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83136]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3552] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83232]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3584] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83328]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3616] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83424]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3648] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87552]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3680] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87648]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3712] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87744]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3744] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87840]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3776] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87936]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3808] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 88032]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3840] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92160]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3872] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92256]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3904] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92352]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3936] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92448]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3968] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92544]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4000] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92640]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4032] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96768]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4064] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96864]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4096] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96960]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4128] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 97056]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4160] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 97152]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4192] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 97248]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4224] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101376]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4256] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101472]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4288] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101568]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4320] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101664]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4352] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101760]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4384] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101856]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4416] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 105984]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4448] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106080]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4480] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106176]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4512] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106272]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4544] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106368]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4576] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106464]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4608] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110592]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4640] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110688]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4672] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110784]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4704] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110880]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4736] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110976]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4768] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 111072]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4800] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115200]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4832] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115296]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4864] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115392]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4896] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115488]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4928] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115584]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4960] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115680]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4992] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 119808]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5024] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 119904]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5056] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120000]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5088] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120096]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5120] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120192]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5152] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120288]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5184] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124416]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5216] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124512]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5248] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124608]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5280] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124704]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5312] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124800]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5344] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124896]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5376] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129024]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5408] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129120]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5440] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129216]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5472] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129312]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5504] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129408]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5536] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129504]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5568] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133632]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5600] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133728]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5632] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133824]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5664] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133920]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5696] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 134016]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5728] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 134112]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5760] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138240]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5792] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138336]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5824] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138432]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5856] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138528]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5888] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138624]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5920] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138720]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5952] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 142848]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5984] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 142944]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6016] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143040]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6048] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143136]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6080] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143232]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6112] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143328]
+ for rc_outer_inner, yy_outer_inner in T.grid(2, 7):
+ cse_var_2: T.int32 = rc_outer_inner * 288 + yy_outer_inner
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 1] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 1]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 2] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 2]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 9] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 3]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 10] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 4]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 11] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 5]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 18] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 6]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 19] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 7]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 20] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 8]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 27] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 9]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 28] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 10]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 29] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 11]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 36] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 12]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 37] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 13]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 38] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 14]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 45] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 15]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 46] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 16]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 47] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 17]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 54] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 18]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 55] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 19]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 56] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 20]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 63] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 21]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 64] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 22]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 65] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 23]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 72] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 24]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 73] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 25]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 74] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 26]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 81] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 27]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 82] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 28]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 83] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 29]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 90] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 30]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 91] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 31]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 92] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 32]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 99] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 33]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 100] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 34]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 101] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 35]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 108] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 36]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 109] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 37]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 110] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 38]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 117] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 39]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 118] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 40]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 119] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 41]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 126] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 42]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 127] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 43]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 128] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 44]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 135] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 45]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 136] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 46]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 137] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 47]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 144] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 48]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 145] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 49]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 146] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 50]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 153] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 51]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 154] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 52]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 155] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 53]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 162] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 54]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 163] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 55]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 164] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 56]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 171] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 57]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 172] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 58]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 173] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 59]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 180] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 60]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 181] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 61]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 182] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 62]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 189] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 63]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 190] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 64]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 191] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 65]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 198] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 66]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 199] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 67]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 200] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 68]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 207] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 69]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 208] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 70]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 209] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 71]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 216] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 72]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 217] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 73]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 218] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 74]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 225] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 75]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 226] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 76]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 227] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 77]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 234] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 78]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 235] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 79]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 236] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 80]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 243] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 81]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 244] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 82]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 245] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 83]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 252] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 84]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 253] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 85]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 254] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 86]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 261] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 87]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 262] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 88]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 263] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 89]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 270] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 90]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 271] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 91]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 272] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 92]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 279] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 93]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 280] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 94]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 281] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 95]
+ for i2_inner in range(7):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x // 7 * 6272 + threadIdx_x * 98 + i1_inner * 49 + blockIdx_x % 7 * 7 + i3_inner] = T.max(conv2d_nchw_1[i1_inner * 7 + i3_inner] + bias_1[blockIdx_x // 7 * 128 + threadIdx_x * 2 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x // 7 * 1568 + threadIdx_x * 49 + i2_inner * 7 + blockIdx_x % 7] = T.max(conv2d_nchw_1[i2_inner] + bias_1[blockIdx_x // 7 * 32 + threadIdx_x], T.float32(0))
@@ -765,7 +835,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.337 ms
+ Execution time of this operator: 0.352 ms
@@ -814,34 +884,34 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
- conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
+ conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=32)
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=1)
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_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_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_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=1)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
- conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=32)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+ conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
- conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
+ conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
+ 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=1)
- compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+ compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, 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_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=1)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
@@ -862,12 +932,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ 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=32)
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=4)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ 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=32)
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, "unroll_explicit", True)
@@ -894,10 +964,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+ extern "C" __global__ void __launch_bounds__(32) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[7];
+ __shared__ float pad_temp_shared[576];
+ __shared__ float kernel_shared[6144];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -905,419 +975,324 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw[4] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- conv2d_nchw[7] = 0.000000e+00f;
- conv2d_nchw[8] = 0.000000e+00f;
- conv2d_nchw[9] = 0.000000e+00f;
- conv2d_nchw[10] = 0.000000e+00f;
- conv2d_nchw[11] = 0.000000e+00f;
- conv2d_nchw[12] = 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) {
+ for (int rc_outer_outer = 0; rc_outer_outer < 8; ++rc_outer_outer) {
+ for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
__syncthreads();
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((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) + 64) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 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) + 128) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 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) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((((((int)threadIdx.x) + 256) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 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) + 320) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 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) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((((((int)threadIdx.x) + 448) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((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) + 512) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 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) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((((((int)threadIdx.x) + 640) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 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) + 704) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 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) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((((((int)threadIdx.x) + 832) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 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) + 896) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((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) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((((((int)threadIdx.x) + 1024) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 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) + 1088) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 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) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((((((int)threadIdx.x) + 1216) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 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) + 1280) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 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) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((((((int)threadIdx.x) + 1408) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 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) + 1472) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 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) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
- kernel_shared[(((((((int)threadIdx.x) + 1600) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 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) + 1664) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 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) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
- kernel_shared[(((((((int)threadIdx.x) + 1792) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 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) + 1856) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 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) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
- kernel_shared[(((((((int)threadIdx.x) + 1984) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 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) + 2048) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 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) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
- kernel_shared[(((((((int)threadIdx.x) + 2176) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 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) + 2240) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 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) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
- kernel_shared[(((((((int)threadIdx.x) + 2368) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 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) + 2432) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 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) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
- kernel_shared[(((((((int)threadIdx.x) + 2560) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 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) + 2624) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 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) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
- kernel_shared[(((((((int)threadIdx.x) + 2752) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 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) + 2816) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 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) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
- kernel_shared[(((((((int)threadIdx.x) + 2944) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 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) + 3008) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((int)threadIdx.x) % 9)) && ((((int)threadIdx.x) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 49)) + ((((int)threadIdx.x) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 32)] = (((((1 <= ((((int)threadIdx.x) + 5) % 9)) && (((((int)threadIdx.x) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 32) / 9) * 49)) + (((((int)threadIdx.x) + 5) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 64)] = (((((1 <= ((((int)threadIdx.x) + 1) % 9)) && (((((int)threadIdx.x) + 1) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 64) / 9) * 49)) + (((((int)threadIdx.x) + 1) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 96)] = (((((1 <= ((((int)threadIdx.x) + 6) % 9)) && (((((int)threadIdx.x) + 6) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 96) / 9) * 49)) + (((((int)threadIdx.x) + 6) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 128)] = (((((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 128) / 9) * 49)) + (((((int)threadIdx.x) + 2) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 160)] = (((((1 <= ((((int)threadIdx.x) + 7) % 9)) && (((((int)threadIdx.x) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 160) / 9) * 49)) + (((((int)threadIdx.x) + 7) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 192)] = (((((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 192) / 9) * 49)) + (((((int)threadIdx.x) + 3) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= ((((int)threadIdx.x) + 8) % 9)) && (((((int)threadIdx.x) + 8) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 224) / 9) * 49)) + (((((int)threadIdx.x) + 8) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 256)] = (((((1 <= ((((int)threadIdx.x) + 4) % 9)) && (((((int)threadIdx.x) + 4) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 256) / 9) * 49)) + (((((int)threadIdx.x) + 4) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 288)] = (((((1 <= (((int)threadIdx.x) % 9)) && ((((int)threadIdx.x) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 49)) + ((((int)threadIdx.x) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) + 1560)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 320)] = (((((1 <= ((((int)threadIdx.x) + 5) % 9)) && (((((int)threadIdx.x) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 320) / 9) * 49)) + (((((int)threadIdx.x) + 5) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 352)] = (((((1 <= ((((int)threadIdx.x) + 1) % 9)) && (((((int)threadIdx.x) + 1) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 352) / 9) * 49)) + (((((int)threadIdx.x) + 1) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 384)] = (((((1 <= ((((int)threadIdx.x) + 6) % 9)) && (((((int)threadIdx.x) + 6) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 384) / 9) * 49)) + (((((int)threadIdx.x) + 6) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 416)] = (((((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 416) / 9) * 49)) + (((((int)threadIdx.x) + 2) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 <= ((((int)threadIdx.x) + 7) % 9)) && (((((int)threadIdx.x) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 448) / 9) * 49)) + (((((int)threadIdx.x) + 7) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 480)] = (((((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 480) / 9) * 49)) + (((((int)threadIdx.x) + 3) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 512)] = (((((1 <= ((((int)threadIdx.x) + 8) % 9)) && (((((int)threadIdx.x) + 8) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 512) / 9) * 49)) + (((((int)threadIdx.x) + 8) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 544)] = (((((1 <= ((((int)threadIdx.x) + 4) % 9)) && (((((int)threadIdx.x) + 4) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 544) / 9) * 49)) + (((((int)threadIdx.x) + 4) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 32)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 64)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 192)];
+ kernel_shared[(((int)threadIdx.x) + 96)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 288)];
+ kernel_shared[(((int)threadIdx.x) + 128)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 384)];
+ kernel_shared[(((int)threadIdx.x) + 160)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 480)];
+ kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4608)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4704)];
+ kernel_shared[(((int)threadIdx.x) + 256)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4800)];
+ kernel_shared[(((int)threadIdx.x) + 288)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4896)];
+ kernel_shared[(((int)threadIdx.x) + 320)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4992)];
+ kernel_shared[(((int)threadIdx.x) + 352)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 5088)];
+ kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9216)];
+ kernel_shared[(((int)threadIdx.x) + 416)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9312)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9408)];
+ kernel_shared[(((int)threadIdx.x) + 480)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9504)];
+ kernel_shared[(((int)threadIdx.x) + 512)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9600)];
+ kernel_shared[(((int)threadIdx.x) + 544)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9696)];
+ kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 13824)];
+ kernel_shared[(((int)threadIdx.x) + 608)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 13920)];
+ kernel_shared[(((int)threadIdx.x) + 640)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14016)];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14112)];
+ kernel_shared[(((int)threadIdx.x) + 704)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14208)];
+ kernel_shared[(((int)threadIdx.x) + 736)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14304)];
+ kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18432)];
+ kernel_shared[(((int)threadIdx.x) + 800)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18528)];
+ kernel_shared[(((int)threadIdx.x) + 832)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18624)];
+ kernel_shared[(((int)threadIdx.x) + 864)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18720)];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18816)];
+ kernel_shared[(((int)threadIdx.x) + 928)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18912)];
+ kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23040)];
+ kernel_shared[(((int)threadIdx.x) + 992)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23136)];
+ kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23232)];
+ kernel_shared[(((int)threadIdx.x) + 1056)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23328)];
+ kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23424)];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23520)];
+ kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27648)];
+ kernel_shared[(((int)threadIdx.x) + 1184)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27744)];
+ kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27840)];
+ kernel_shared[(((int)threadIdx.x) + 1248)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27936)];
+ kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 28032)];
+ kernel_shared[(((int)threadIdx.x) + 1312)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 28128)];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 1376)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32352)];
+ kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32448)];
+ kernel_shared[(((int)threadIdx.x) + 1440)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32544)];
+ kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32640)];
+ kernel_shared[(((int)threadIdx.x) + 1504)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32736)];
+ kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 36864)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 36960)];
+ kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37056)];
+ kernel_shared[(((int)threadIdx.x) + 1632)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37152)];
+ kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37248)];
+ kernel_shared[(((int)threadIdx.x) + 1696)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37344)];
+ kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41472)];
+ kernel_shared[(((int)threadIdx.x) + 1760)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41568)];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41664)];
+ kernel_shared[(((int)threadIdx.x) + 1824)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41760)];
+ kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41856)];
+ kernel_shared[(((int)threadIdx.x) + 1888)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41952)];
+ kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46080)];
+ kernel_shared[(((int)threadIdx.x) + 1952)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46176)];
+ kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46272)];
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46368)];
+ kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46464)];
+ kernel_shared[(((int)threadIdx.x) + 2080)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46560)];
+ kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50688)];
+ kernel_shared[(((int)threadIdx.x) + 2144)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50784)];
+ kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50880)];
+ kernel_shared[(((int)threadIdx.x) + 2208)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50976)];
+ kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 51072)];
+ kernel_shared[(((int)threadIdx.x) + 2272)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 51168)];
+ kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55296)];
+ kernel_shared[(((int)threadIdx.x) + 2336)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55392)];
+ kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55488)];
+ kernel_shared[(((int)threadIdx.x) + 2400)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55584)];
+ kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55680)];
+ kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55776)];
+ kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 59904)];
+ kernel_shared[(((int)threadIdx.x) + 2528)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60000)];
+ kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60096)];
+ kernel_shared[(((int)threadIdx.x) + 2592)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60192)];
+ kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60288)];
+ kernel_shared[(((int)threadIdx.x) + 2656)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60384)];
+ kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 2720)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64608)];
+ kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64704)];
+ kernel_shared[(((int)threadIdx.x) + 2784)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64800)];
+ kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64896)];
+ kernel_shared[(((int)threadIdx.x) + 2848)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64992)];
+ kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69120)];
+ kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69216)];
+ kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69312)];
+ kernel_shared[(((int)threadIdx.x) + 2976)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69408)];
+ kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69504)];
+ kernel_shared[(((int)threadIdx.x) + 3040)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69600)];
+ kernel_shared[(((int)threadIdx.x) + 3072)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 73728)];
+ kernel_shared[(((int)threadIdx.x) + 3104)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 73824)];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 73920)];
+ kernel_shared[(((int)threadIdx.x) + 3168)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 74016)];
+ kernel_shared[(((int)threadIdx.x) + 3200)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 74112)];
+ kernel_shared[(((int)threadIdx.x) + 3232)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 74208)];
+ kernel_shared[(((int)threadIdx.x) + 3264)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78336)];
+ kernel_shared[(((int)threadIdx.x) + 3296)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78432)];
+ kernel_shared[(((int)threadIdx.x) + 3328)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78528)];
+ kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78624)];
+ kernel_shared[(((int)threadIdx.x) + 3392)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78720)];
+ kernel_shared[(((int)threadIdx.x) + 3424)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78816)];
+ kernel_shared[(((int)threadIdx.x) + 3456)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 82944)];
+ kernel_shared[(((int)threadIdx.x) + 3488)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83040)];
+ kernel_shared[(((int)threadIdx.x) + 3520)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83136)];
+ kernel_shared[(((int)threadIdx.x) + 3552)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83232)];
+ kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83328)];
+ kernel_shared[(((int)threadIdx.x) + 3616)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83424)];
+ kernel_shared[(((int)threadIdx.x) + 3648)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87552)];
+ kernel_shared[(((int)threadIdx.x) + 3680)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87648)];
+ kernel_shared[(((int)threadIdx.x) + 3712)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87744)];
+ kernel_shared[(((int)threadIdx.x) + 3744)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87840)];
+ kernel_shared[(((int)threadIdx.x) + 3776)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87936)];
+ kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 88032)];
+ kernel_shared[(((int)threadIdx.x) + 3840)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92160)];
+ kernel_shared[(((int)threadIdx.x) + 3872)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92256)];
+ kernel_shared[(((int)threadIdx.x) + 3904)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92352)];
+ kernel_shared[(((int)threadIdx.x) + 3936)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92448)];
+ kernel_shared[(((int)threadIdx.x) + 3968)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92544)];
+ kernel_shared[(((int)threadIdx.x) + 4000)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92640)];
+ kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96768)];
+ kernel_shared[(((int)threadIdx.x) + 4064)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96864)];
+ kernel_shared[(((int)threadIdx.x) + 4096)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96960)];
+ kernel_shared[(((int)threadIdx.x) + 4128)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 97056)];
+ kernel_shared[(((int)threadIdx.x) + 4160)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 97152)];
+ kernel_shared[(((int)threadIdx.x) + 4192)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 97248)];
+ kernel_shared[(((int)threadIdx.x) + 4224)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101376)];
+ kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101472)];
+ kernel_shared[(((int)threadIdx.x) + 4288)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101568)];
+ kernel_shared[(((int)threadIdx.x) + 4320)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101664)];
+ kernel_shared[(((int)threadIdx.x) + 4352)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101760)];
+ kernel_shared[(((int)threadIdx.x) + 4384)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101856)];
+ kernel_shared[(((int)threadIdx.x) + 4416)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 105984)];
+ kernel_shared[(((int)threadIdx.x) + 4448)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106080)];
+ kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106176)];
+ kernel_shared[(((int)threadIdx.x) + 4512)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106272)];
+ kernel_shared[(((int)threadIdx.x) + 4544)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106368)];
+ kernel_shared[(((int)threadIdx.x) + 4576)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106464)];
+ kernel_shared[(((int)threadIdx.x) + 4608)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110592)];
+ kernel_shared[(((int)threadIdx.x) + 4640)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110688)];
+ kernel_shared[(((int)threadIdx.x) + 4672)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110784)];
+ kernel_shared[(((int)threadIdx.x) + 4704)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110880)];
+ kernel_shared[(((int)threadIdx.x) + 4736)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110976)];
+ kernel_shared[(((int)threadIdx.x) + 4768)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 111072)];
+ kernel_shared[(((int)threadIdx.x) + 4800)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115200)];
+ kernel_shared[(((int)threadIdx.x) + 4832)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115296)];
+ kernel_shared[(((int)threadIdx.x) + 4864)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115392)];
+ kernel_shared[(((int)threadIdx.x) + 4896)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115488)];
+ kernel_shared[(((int)threadIdx.x) + 4928)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115584)];
+ kernel_shared[(((int)threadIdx.x) + 4960)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115680)];
+ kernel_shared[(((int)threadIdx.x) + 4992)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 119808)];
+ kernel_shared[(((int)threadIdx.x) + 5024)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 119904)];
+ kernel_shared[(((int)threadIdx.x) + 5056)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120000)];
+ kernel_shared[(((int)threadIdx.x) + 5088)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120096)];
+ kernel_shared[(((int)threadIdx.x) + 5120)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120192)];
+ kernel_shared[(((int)threadIdx.x) + 5152)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120288)];
+ kernel_shared[(((int)threadIdx.x) + 5184)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124416)];
+ kernel_shared[(((int)threadIdx.x) + 5216)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124512)];
+ kernel_shared[(((int)threadIdx.x) + 5248)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124608)];
+ kernel_shared[(((int)threadIdx.x) + 5280)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124704)];
+ kernel_shared[(((int)threadIdx.x) + 5312)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124800)];
+ kernel_shared[(((int)threadIdx.x) + 5344)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124896)];
+ kernel_shared[(((int)threadIdx.x) + 5376)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129024)];
+ kernel_shared[(((int)threadIdx.x) + 5408)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129120)];
+ kernel_shared[(((int)threadIdx.x) + 5440)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129216)];
+ kernel_shared[(((int)threadIdx.x) + 5472)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129312)];
+ kernel_shared[(((int)threadIdx.x) + 5504)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129408)];
+ kernel_shared[(((int)threadIdx.x) + 5536)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129504)];
+ kernel_shared[(((int)threadIdx.x) + 5568)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133632)];
+ kernel_shared[(((int)threadIdx.x) + 5600)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133728)];
+ kernel_shared[(((int)threadIdx.x) + 5632)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133824)];
+ kernel_shared[(((int)threadIdx.x) + 5664)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133920)];
+ kernel_shared[(((int)threadIdx.x) + 5696)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 134016)];
+ kernel_shared[(((int)threadIdx.x) + 5728)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 134112)];
+ kernel_shared[(((int)threadIdx.x) + 5760)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138240)];
+ kernel_shared[(((int)threadIdx.x) + 5792)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138336)];
+ kernel_shared[(((int)threadIdx.x) + 5824)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138432)];
+ kernel_shared[(((int)threadIdx.x) + 5856)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138528)];
+ kernel_shared[(((int)threadIdx.x) + 5888)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138624)];
+ kernel_shared[(((int)threadIdx.x) + 5920)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138720)];
+ kernel_shared[(((int)threadIdx.x) + 5952)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 142848)];
+ kernel_shared[(((int)threadIdx.x) + 5984)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 142944)];
+ kernel_shared[(((int)threadIdx.x) + 6016)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143040)];
+ kernel_shared[(((int)threadIdx.x) + 6048)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143136)];
+ kernel_shared[(((int)threadIdx.x) + 6080)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143232)];
+ kernel_shared[(((int)threadIdx.x) + 6112)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143328)];
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 2; ++rc_outer_inner) {
+ for (int yy_outer_inner = 0; yy_outer_inner < 7; ++yy_outer_inner) {
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[((rc_outer_inner * 288) + yy_outer_inner)] * kernel_shared[((((int)threadIdx.x) * 192) + (rc_outer_inner * 96))]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 1)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 1)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 2)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 2)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 9)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 3)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 10)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 4)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 11)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 5)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 18)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 6)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 19)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 7)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 20)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 8)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 27)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 9)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 28)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 10)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 29)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 11)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 36)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 12)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 37)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 13)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 38)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 14)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 45)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 15)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 46)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 16)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 47)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 17)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 54)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 18)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 55)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 19)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 56)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 20)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 63)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 21)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 64)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 22)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 65)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 23)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 72)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 24)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 73)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 25)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 74)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 26)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 81)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 27)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 82)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 28)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 83)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 29)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 90)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 30)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 91)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 31)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 92)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 32)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 99)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 33)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 100)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 34)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 101)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 35)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 108)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 36)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 109)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 37)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 110)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 38)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 117)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 39)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 118)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 40)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 119)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 41)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 126)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 42)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 127)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 43)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 128)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 44)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 135)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 45)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 136)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 46)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 137)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 47)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 144)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 48)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 145)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 49)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 146)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 50)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 153)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 51)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 154)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 52)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 155)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 53)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 162)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 54)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 163)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 55)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 164)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 56)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 171)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 57)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 172)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 58)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 173)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 59)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 180)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 60)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 181)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 61)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 182)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 62)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 189)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 63)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 190)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 64)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 191)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 65)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 198)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 66)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 199)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 67)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 200)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 68)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 207)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 69)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 208)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 70)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 209)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 71)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 216)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 72)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 217)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 73)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 218)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 74)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 225)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 75)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 226)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 76)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 227)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 77)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 234)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 78)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 235)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 79)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 236)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 80)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 243)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 81)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 244)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 82)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 245)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 83)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 252)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 84)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 253)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 85)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 254)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 86)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 261)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 87)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 262)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 88)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 263)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 89)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 270)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 90)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 271)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 91)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 272)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 92)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 279)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 93)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 280)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 94)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 281)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 95)]));
+ }
+ }
}
}
- for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
- }
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) / 7) * 1568) + (((int)threadIdx.x) * 49)) + (i2_inner * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[(((((int)blockIdx.x) / 7) * 32) + ((int)threadIdx.x))]), 0.000000e+00f);
}
}
@@ -1377,7 +1352,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 6 minutes 19.519 seconds)
+ **Total running time of the script:** ( 6 minutes 45.477 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 c7fe4af329..fba047758d 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -647,7 +647,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1154 8.1176 8.1183 8.1103 0.0036
+ 8.1331 8.1342 8.1374 8.1277 0.0040
@@ -675,7 +675,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 13.644 seconds)
+ **Total running time of the script:** ( 1 minutes 16.077 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 a18f8185f1..22514fcf37 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)
- 781.4774 768.0435 809.1578 767.2309 19.5758
+ 794.8575 794.8890 796.3091 793.3742 1.1984
@@ -694,7 +694,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 46.122 seconds)
+ **Total running time of the script:** ( 1 minutes 49.529 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 2db7a80027..5b37127adb 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -389,26 +389,86 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- for i0_outer_i1_outer_fused in T.parallel(64):
- compute_1 = T.allocate([1024], "float32", "global")
- compute_2 = T.Buffer((1024,), data=compute_1)
- for nb_j_inner in range(2):
- for i_inner_init, j_init in T.grid(32, 16):
- compute_2[i_inner_init * 32 + nb_j_inner * 16 + j_init] = T.float32(0)
- for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner}), 32, 16):
- cse_var_1 = T.int32()
- placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_3: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
- cse_var_2: T.int32 = i_inner * 32 + nb_j_inner * 16 + j
- placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
- placeholder_7 = T.Buffer((32768,), data=placeholder.data)
- placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 8192 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- for i0_inner in range(32):
- cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 16384 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32
- compute_3 = T.Buffer((65536,), data=compute.data)
- placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_4:cse_var_4 + 32] = T.max(compute_2[i0_inner * 32:i0_inner * 32 + 32] + placeholder_5[cse_var_4:cse_var_4 + 32], T.Broadcast(T.float32(0), 32))
+ for i0_outer in T.parallel(32):
+ compute_1 = T.allocate([32], "float32", "global")
+ for i1_outer in range(64):
+ compute_2 = T.Buffer((32,), data=compute_1)
+ for i_outer_inner in range(2):
+ cse_var_2: T.int32 = i1_outer // 2
+ cse_var_1: T.int32 = i_outer_inner * 16
+ compute_2[cse_var_1] = T.float32(0)
+ compute_2[cse_var_1 + 1] = T.float32(0)
+ compute_2[cse_var_1 + 2] = T.float32(0)
+ compute_2[cse_var_1 + 3] = T.float32(0)
+ compute_2[cse_var_1 + 4] = T.float32(0)
+ compute_2[cse_var_1 + 5] = T.float32(0)
+ compute_2[cse_var_1 + 6] = T.float32(0)
+ compute_2[cse_var_1 + 7] = T.float32(0)
+ compute_2[cse_var_1 + 8] = T.float32(0)
+ compute_2[cse_var_1 + 9] = T.float32(0)
+ compute_2[cse_var_1 + 10] = T.float32(0)
+ compute_2[cse_var_1 + 11] = T.float32(0)
+ compute_2[cse_var_1 + 12] = T.float32(0)
+ compute_2[cse_var_1 + 13] = T.float32(0)
+ compute_2[cse_var_1 + 14] = T.float32(0)
+ compute_2[cse_var_1 + 15] = T.float32(0)
+ for elem_idx in range(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
+ placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
+ placeholder_7 = T.Buffer((32768,), data=placeholder.data)
+ placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ compute_2[cse_var_1] = compute_2[cse_var_1] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_3: T.int32 = cse_var_1 + 1
+ compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_4: T.int32 = cse_var_1 + 2
+ compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_5: T.int32 = cse_var_1 + 3
+ compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_6: T.int32 = cse_var_1 + 4
+ compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_7: T.int32 = cse_var_1 + 5
+ compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_8: T.int32 = cse_var_1 + 6
+ compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_9: T.int32 = cse_var_1 + 7
+ compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_10: T.int32 = cse_var_1 + 8
+ compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_11: T.int32 = cse_var_1 + 9
+ compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_12: T.int32 = cse_var_1 + 10
+ compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_13: T.int32 = cse_var_1 + 11
+ compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_14: T.int32 = cse_var_1 + 12
+ compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_15: T.int32 = cse_var_1 + 13
+ compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_16: T.int32 = cse_var_1 + 14
+ compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_17: T.int32 = cse_var_1 + 15
+ compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ for i0_inner in range(4):
+ cse_var_18: T.int32 = i0_outer * 2048 + i0_inner * 512 + i1_outer * 8
+ compute_3 = T.Buffer((65536,), data=compute.data)
+ placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
+ compute_3[cse_var_18:cse_var_18 + 8] = T.max(compute_2[i0_inner * 8:i0_inner * 8 + 8] + placeholder_5[cse_var_18:cse_var_18 + 8], T.Broadcast(T.float32(0), 8))
@@ -458,7 +518,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.644 ms
+ Execution time of this operator: 2.009 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 bc725305de..6a77a19192 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
=================
-**00:37.283** total execution time for **how_to_tune_with_autotvm** files:
+**00:50.982** 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``) | 00:37.246 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:50.944 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.022 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.023 | 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_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.005 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``) | 00:00.006 | 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``) | 00:00.005 | 0.0 MB |
++--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index f524a2c1bd..0cf01f3818 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
@@ -390,254 +390,163 @@ 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, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1310820
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3196677
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)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "/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 399, 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:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../src/runtime/rpc/rpc_module.cc:129
+ 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1012
+ 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
+ at ../src/runtime/rpc/rpc_endpoint.cc:804
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 804
+ TVMError:
+ ---------------------------------------------------------------
+ An error occurred during the execution of TVM.
+ For more information, please see: https://tvm.apache.org/docs/errors.html
+ ---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
+
+ During handling of the above exception, another exception occurred:
Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10019580
- 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
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
- File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
- File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
- File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+ 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 179, 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):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ 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: 0x00007fa543335fa2
+ 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:187
+ 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:1621
2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
at ../include/tvm/runtime/packed_func.h:1217
1: Call
at ../include/tvm/runtime/packed_func.h:1213
0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 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):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10188163
- No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ 52: 0xffffffffffffffff
+ 51: _start
+ 50: __libc_start_main
+ 49: _Py_UnixMain
+ 48: 0x0000000000650da0
+ 47: 0x0000000000650afa
+ 46: _PyFunction_FastCallDict
+ 45: _PyEval_EvalCodeWithName
+ 44: _PyEval_EvalFrameDefault
+ 43: _PyFunction_FastCallKeywords
+ 42: _PyEval_EvalCodeWithName
+ 41: _PyEval_EvalFrameDefault
+ 40: _PyMethodDef_RawFastCallKeywords
+ 39: 0x0000000000546369
+ 38: _PyEval_EvalCodeWithName
+ 37: _PyEval_EvalFrameDefault
+ 36: _PyFunction_FastCallKeywords
+ 35: _PyEval_EvalCodeWithName
+ 34: _PyEval_EvalFrameDefault
+ 33: _PyFunction_FastCallDict
+ 32: _PyEval_EvalCodeWithName
+ 31: _PyEval_EvalFrameDefault
+ 30: _PyObject_FastCallDict
+ 29: 0x00000000004c06e1
+ 28: _PyFunction_FastCallDict
+ 27: _PyEval_EvalFrameDefault
+ 26: _PyMethodDescr_FastCallKeywords
+ 25: 0x00000000005dcb58
+ 24: 0x00000000005dc83f
+ 23: 0x00000000004ba127
+ 22: _PyEval_EvalFrameDefault
+ 21: _PyFunction_FastCallKeywords
+ 20: _PyEval_EvalFrameDefault
+ 19: _PyFunction_FastCall [('tile_f', [-1, 32, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3749161
+ No: 3 GFLOPS: 11.81/11.81 result: MeasureResult(costs=(0.019608032166666668,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.010102033615112, timestamp=1682098553.3948522) [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8596601
+ No: 4 GFLOPS: 239.17/239.17 result: MeasureResult(costs=(0.0009679504098360656,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1777567863464355, timestamp=1682098556.048608) [('tile_f', [-1, 2, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8233975
+ No: 5 GFLOPS: 0.00/239.17 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
@@ -759,8 +668,8 @@ 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, 256, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3777848
- No: 5 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 512]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1411299
+ No: 6 GFLOPS: 0.00/239.17 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
@@ -882,10 +791,8 @@ 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, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5887318
- No: 6 GFLOPS: 25.87/25.87 result: MeasureResult(costs=(0.0089472535,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.663716316223145, timestamp=1682041470.0681114) [('tile_f', [-1, 32, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3884397
- No: 7 GFLOPS: 5.97/25.87 result: MeasureResult(costs=(0.03877114475,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3159408569335938, timestamp=1682041471.1311076) [('tile_f', [-1, 1, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2080550
- No: 8 GFLOPS: 0.00/25.87 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9545468
+ No: 7 GFLOPS: 0.00/239.17 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
@@ -1007,9 +914,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1437864
- No: 9 GFLOPS: 70.75/70.75 result: MeasureResult(costs=(0.0032721330526315788,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4330008029937744, timestamp=1682041473.2224634) [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1809759
- No: 10 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9735445
+ No: 8 GFLOPS: 0.00/239.17 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
@@ -1131,8 +1037,10 @@ 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, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4255353
- No: 11 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7702724
+ No: 9 GFLOPS: 94.19/239.17 result: MeasureResult(costs=(0.0024577285599999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9577372074127197, timestamp=1682098563.4863958) [('tile_f', [-1, 16, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4852554
+ No: 10 GFLOPS: 30.39/239.17 result: MeasureResult(costs=(0.007618267571428571,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.270705223083496, timestamp=1682098564.4030895) [('tile_f', [-1, 16, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9521896
+ No: 11 GFLOPS: 0.00/239.17 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
@@ -1254,9 +1162,9 @@ 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, 1, 32, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4249560
- No: 12 GFLOPS: 26.73/70.75 result: MeasureResult(costs=(0.00866072775,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8893969058990479, timestamp=1682041474.1282947) [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,107745
- No: 13 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8679379
+ No: 12 GFLOPS: 94.16/239.17 result: MeasureResult(costs=(0.002458553024390244,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6427075862884521, timestamp=1682098565.275473) [('tile_f', [-1, 1, 64, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1940563
+ No: 13 GFLOPS: 0.00/239.17 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
@@ -1378,131 +1286,161 @@ 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, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7949889
- No: 14 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
- func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 2, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1893611
+ No: 14 GFLOPS: 0.00/239.17 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 399, 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:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../src/runtime/rpc/rpc_module.cc:129
+ 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1012
+ 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
+ at ../src/runtime/rpc/rpc_endpoint.cc:804
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 804
+ TVMError:
+ ---------------------------------------------------------------
+ An error occurred during the execution of TVM.
+ For more information, please see: https://tvm.apache.org/docs/errors.html
+ ---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
+
+ During handling of the above exception, another exception occurred:
Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ 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 179, 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: 0x00007fa6122a5fa2
+ 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:187
+ 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:1621
2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
at ../include/tvm/runtime/packed_func.h:1217
1: Call
at ../include/tvm/runtime/packed_func.h:1213
0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4636507
- No: 15 GFLOPS: 0.00/70.75 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, 4, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5619662
+ No: 15 GFLOPS: 0.00/239.17 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
@@ -1624,8 +1562,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 128, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9836689
- No: 16 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1771785
+ No: 16 GFLOPS: 0.00/239.17 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
@@ -1747,8 +1685,8 @@ 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, 8, 64]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,307769
- No: 17 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9908733
+ No: 17 GFLOPS: 0.00/239.17 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
@@ -1870,8 +1808,8 @@ 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, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6531548
- No: 18 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5735812
+ No: 18 GFLOPS: 0.00/239.17 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
@@ -1993,8 +1931,8 @@ 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, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7579363
- No: 19 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9485620
+ No: 19 GFLOPS: 0.00/239.17 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
@@ -2116,8 +2054,8 @@ 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, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4626423
- No: 20 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6724763
+ No: 20 GFLOPS: 0.00/239.17 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
@@ -2239,7 +2177,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,333279
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9774217
@@ -2294,9 +2232,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1809759
+ [('tile_f', [-1, 2, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8233975
Finish loading 20 records
- Time cost of this operator: 0.002993
+ Time cost of this operator: 0.001260
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 d0fc8667c3..820fba5fd1 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -360,10 +360,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 315.6 98.667 (1, 2, 10, 10, 3) 2 1 [315.6]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.312 1.035 (1, 6, 10, 10) 1 1 [3.312]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.952 0.298 (1, 1, 10, 10, 3) 1 1 [0.952]
- Total_time - 319.864 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 317.2 98.743 (1, 2, 10, 10, 3) 2 1 [317.2]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.067 0.955 (1, 6, 10, 10) 1 1 [3.067]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.971 0.302 (1, 1, 10, 10, 3) 1 1 [0.971]
+ Total_time - 321.238 - - - - -
@@ -428,10 +428,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 101.4 97.41 (1, 6, 10, 10, 1) 2 1 [101.4]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.74 1.672 (1, 6, 10, 10) 1 1 [1.74]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.956 0.918 (1, 1, 10, 10, 3) 1 1 [0.956]
- Total_time - 104.096 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 103.5 97.516 (1, 6, 10, 10, 1) 2 1 [103.5]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.793 1.689 (1, 6, 10, 10) 1 1 [1.793]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.843 0.795 (1, 3, 10, 10, 1) 1 1 [0.843]
+ Total_time - 106.136 - - - - -
@@ -439,7 +439,7 @@ Timing the tuned program
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 27.079 seconds)
+ **Total running time of the script:** ( 1 minutes 32.563 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_autotune.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
index a2baa7ca11..cc02bbf830 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
@@ -118,7 +118,7 @@ download a cat image and preprocess it to use as the model input.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/ao/quantization/utils.py:281: UserWarning: must run observer before calling calculate_qparams. Returning default values.
"must run observer before calling calculate_qparams. " +
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
-
0%| | 0.00/3.42M [00:00<?, ?B/s]
61%|###### | 2.09M/3.42M [00:00<00:00, 12.4MB/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 19.7MB/s]
+
0%| | 0.00/3.42M [00:00<?, ?B/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 92.2MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -324,7 +324,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 21.805 seconds)
+ **Total running time of the script:** ( 1 minutes 25.799 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_pytorch.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
index 4ac617a698..8d55ae6a08 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -217,7 +217,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpqjisivml/images/random'
+ '/tmp/tmp1db7xoih/images/random'
@@ -308,7 +308,7 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. image-sg:: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
- :alt: [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0]
+ :alt: [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -317,8 +317,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpqjisivml/images/target contains 8144 images
- /tmp/tmpqjisivml/images/random contains 5000 images
+ /tmp/tmp1db7xoih/images/target contains 8144 images
+ /tmp/tmp1db7xoih/images/random contains 5000 images
@@ -493,13 +493,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 42s - loss: 0.2136 - accuracy: 0.9228 - val_loss: 0.1013 - val_accuracy: 0.9622 - 42s/epoch - 129ms/step
+ 328/328 - 41s - loss: 0.2278 - accuracy: 0.9230 - val_loss: 0.1670 - val_accuracy: 0.9430 - 41s/epoch - 125ms/step
Epoch 2/3
- 328/328 - 35s - loss: 0.0958 - accuracy: 0.9647 - val_loss: 0.1078 - val_accuracy: 0.9622 - 35s/epoch - 107ms/step
+ 328/328 - 35s - loss: 0.0970 - accuracy: 0.9642 - val_loss: 0.1137 - val_accuracy: 0.9607 - 35s/epoch - 108ms/step
Epoch 3/3
- 328/328 - 35s - loss: 0.0707 - accuracy: 0.9745 - val_loss: 0.0993 - val_accuracy: 0.9668 - 35s/epoch - 107ms/step
+ 328/328 - 35s - loss: 0.0717 - accuracy: 0.9730 - val_loss: 0.1333 - val_accuracy: 0.9664 - 35s/epoch - 108ms/step
- <keras.callbacks.History object at 0x7fb847364fd0>
+ <keras.callbacks.History object at 0x7ffa057aea10>
@@ -860,7 +860,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 24.422 seconds)
+ **Total running time of the script:** ( 4 minutes 32.404 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 274b6fbf19..64c568d6a1 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,20 +5,20 @@
Computation times
=================
-**07:40.221** total execution time for **how_to_work_with_microtvm** files:
+**07:59.150** total execution time for **how_to_work_with_microtvm** files:
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:24.422 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:32.404 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:27.079 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:32.563 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:21.805 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:25.799 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.761 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:11.252 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:08.372 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:09.031 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.782 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:08.102 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.000 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index d96f2879ea..1c080faffa 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
=================
-**00:38.105** total execution time for **how_to_work_with_relay** files:
+**00:39.848** 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``) | 00:33.375 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:35.046 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:03.015 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:03.109 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.709 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.685 | 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 |
+| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.007 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index 6021615005..f86ad85b73 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
@@ -278,7 +278,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7fb6f0170b00>
+ <function my_cuda_math_rule at 0x7ff79e34f7a0>
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 4aabda905b..703099a891 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,22 +5,22 @@
Computation times
=================
-**00:05.591** total execution time for **how_to_work_with_schedules** files:
+**00:09.451** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:02.693 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:06.559 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.328 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.277 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.652 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.664 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.631 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.643 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.134 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.142 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.063 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.068 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.058 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.063 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.031 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.034 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index b85baceaa0..9796288b76 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
=================
-**00:32.815** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:34.025** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:32.808 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:34.018 | 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 b537f35427..f139b76242 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 35.72s!
+ resnet18_v1 inference graph built in 36.92s!
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 2999d8d7c0..4aeb021b12 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -337,7 +337,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 23.96s!
+ yolov3-tiny inference graph built in 24.89s!
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 7a1d56bba3..bbdc0e5169 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
=================
-**01:43.882** total execution time for **topic_vta_tutorials_frontend** files:
+**01:46.342** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:52.735 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:54.015 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:51.146 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:52.327 | 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 9dcd908c71..3e65b5e1d7 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
=================
-**00:03.260** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.281** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.747 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.770 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.513 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.511 | 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 7cb646cf9b..8cc63973c9 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
=================
-**00:00.886** total execution time for **topic_vta_tutorials** files:
+**00:00.880** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.464 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.455 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.422 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.424 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index 3e64f136cd..56122d1919 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
+
@@ -318,7 +325,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 95.118 ms
+ Execution time of this operator: 101.593 ms
@@ -434,7 +441,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 28.212 seconds)
+ **Total running time of the script:** ( 1 minutes 37.901 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 70cc873508..c630e07b07 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: 1.54/1.54 result: MeasureResult(costs=(0.174234263,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.027167797088623, timestamp=1682039737.3193986) [('tile_y', [-1, 1]), ('tile_x', [-1, 1])],None,0
- No: 2 GFLOPS: 0.85/1.54 result: MeasureResult(costs=(0.31709130599999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.314919710159302, timestamp=1682039742.656767) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
- No: 3 GFLOPS: 10.09/10.09 result: MeasureResult(costs=(0.026598257999999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7162415981292725, timestamp=1682039744.6474473) [('tile_y', [-1, 16]), ('tile_x', [-1, 32])],None,54
- No: 4 GFLOPS: 1.95/10.09 result: MeasureResult(costs=(0.137371136,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4194116592407227, timestamp=1682039748.4069366) [('tile_y', [-1, 8]), ('tile_x', [-1, 1])],None,3
- No: 5 GFLOPS: 8.71/10.09 result: MeasureResult(costs=(0.0308292648,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7429296970367432, timestamp=1682039750.5717063) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
- No: 6 GFLOPS: 0.89/10.09 result: MeasureResult(costs=(0.2999806986,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.056335210800171, timestamp=1682039755.6349726) [('tile_y', [-1, 64]), ('tile_x', [-1, 2])],None,16
- No: 7 GFLOPS: 0.95/10.09 result: MeasureResult(costs=(0.2834529414,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.791394472122192, timestamp=1682039760.4350102) [('tile_y', [-1, 32]), ('tile_x', [-1, 2])],None,15
- No: 8 GFLOPS: 3.43/10.09 result: MeasureResult(costs=(0.0783356682,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5101521015167236, timestamp=1682039761.9555604) [('tile_y', [-1, 8]), ('tile_x', [-1, 8])],None,33
- No: 9 GFLOPS: 8.86/10.09 result: MeasureResult(costs=(0.030288893,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7007701396942139, timestamp=1682039762.7863314) [('tile_y', [-1, 512]), ('tile_x', [-1, 512])],None,99
- No: 10 GFLOPS: 10.21/10.21 result: MeasureResult(costs=(0.026283970400000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6476967334747314, timestamp=1682039763.4742224) [('tile_y', [-1, 256]), ('tile_x', [-1, 256])],None,88
+ No: 1 GFLOPS: 10.18/10.18 result: MeasureResult(costs=(0.026377261199999995,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6973414421081543, timestamp=1682096747.6508598) [('tile_y', [-1, 256]), ('tile_x', [-1, 32])],None,58
+ No: 2 GFLOPS: 2.12/10.18 result: MeasureResult(costs=(0.12690639080000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.300227165222168, timestamp=1682096751.3316772) [('tile_y', [-1, 8]), ('tile_x', [-1, 2])],None,13
+ No: 3 GFLOPS: 1.58/10.18 result: MeasureResult(costs=(0.169804233,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.9775302410125732, timestamp=1682096755.6928508) [('tile_y', [-1, 32]), ('tile_x', [-1, 4])],None,25
+ No: 4 GFLOPS: 1.60/10.18 result: MeasureResult(costs=(0.1678691298,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.9236934185028076, timestamp=1682096760.0139651) [('tile_y', [-1, 64]), ('tile_x', [-1, 4])],None,26
+ No: 5 GFLOPS: 12.40/12.40 result: MeasureResult(costs=(0.021647650799999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.619530200958252, timestamp=1682096760.7788258) [('tile_y', [-1, 256]), ('tile_x', [-1, 64])],None,68
+ No: 6 GFLOPS: 6.76/12.40 result: MeasureResult(costs=(0.039726912399999995,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.9388878345489502, timestamp=1682096761.6880138) [('tile_y', [-1, 512]), ('tile_x', [-1, 128])],None,79
+ No: 7 GFLOPS: 7.60/12.40 result: MeasureResult(costs=(0.035315636,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8220257759094238, timestamp=1682096762.525237) [('tile_y', [-1, 1]), ('tile_x', [-1, 32])],None,50
+ No: 8 GFLOPS: 2.08/12.40 result: MeasureResult(costs=(0.129281467,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3142731189727783, timestamp=1682096764.859578) [('tile_y', [-1, 4]), ('tile_x', [-1, 1])],None,2
+ No: 9 GFLOPS: 2.31/12.40 result: MeasureResult(costs=(0.11637837799999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.086926221847534, timestamp=1682096767.079089) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
+ No: 10 GFLOPS: 2.89/12.40 result: MeasureResult(costs=(0.0930362492,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7288262844085693, timestamp=1682096768.8401315) [('tile_y', [-1, 8]), ('tile_x', [-1, 4])],None,23
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index c5323fb4ce..928f17efe9 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': 513.6553296399961, 'median': 513.3987593500024, 'std': 1.727023582978912}
+ {'mean': 524.1644680799982, 'median': 524.1663549999998, 'std': 0.5732194644729002}
@@ -582,31 +582,31 @@ 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: 8.88/ 17.38 GFLOPS | Progress: (4/20) | 12.66 s
[Task 1/25] Current/Best: 14.50/ 17.95 GFLOPS | Progress: (8/20) | 17.10 s
[Task 1/25] Current/Best: 8.61/ 17.95 GFLOPS | Progress: (12/20) | 21.02 s
[Task 1/25] Current/Best: 17.60/ 17.95 GFLOPS | Progress: (16/20) | 24.30 s
[Task 1/25] Current/Best: 9.94/ 17.95 GFLOPS | Progress: (20/20) | 28.23 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.78/ 17.97 GFLOPS | Progress: (4/20) | 4.99 s
[Task 2/25] Current/Best: 9.59/ 17.97 GFLOPS | Progress: (8/20) | 6.61 s
[Task 2/25] Current/Best: 8.59/ 17.97 GFLOPS | Progress: (12/20) | 8.42 s
[Task 2/25] Current/Best: 17.88/ 17.97 GFLOPS | Progress: (16/20) | 10.06 s
[Task 2/25] Current/Best: 6.45/ 17.97 GFLOPS | Progress: (20/20) | 12.16 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 15.38/ 19.73 GFLOPS | Progress: (4/20) | 4.86 s
[Task 3/25] Current/Best: 12.41/ 19.73 GFLOPS | Progress: (8/20) | 7.17 s
[Task 3/25] Current/Best: 6.99/ 23.04 GFLOPS | Progress: (12/20) | 9.90 s
[Task 3/25] Current/Best: 11.93/ 23.04 GFLOPS | Progress: (16/20) | 12.69 s
[Task 3/25] Current/Best: 7.04/ 23.04 GFLOPS | Progress: (20/20) | 14.91 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 8.30/ 16.08 GFLOPS | Progress: (4/20) | 13.99 s
[Task 4/25] Current/Best: 19.46/ 19.46 GFLOPS | Progress: (8/20) | 22.42 s
[Task 4/25] Current/Best: 10.94/ 19.46 GFLOPS | Progress: (12/20) | 25.12 s
[Task 4/25] Current/Best: 10.60/ 22.34 GFLOPS | Progress: (16/20) | 27.42 s
[Task 4/25] Current/Best: 8.32/ 22.34 GFLOPS | Progress: (20/20) | 32.92 s
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 12.42/ 17.33 GFLOPS | Progress: (4/20) | 5.55 s
[Task 5/25] Current/Best: 5.03/ 17.33 GFLOPS | Progress: (8/20) | 7.50 s Done.
-
[Task 5/25] Current/Best: 13.24/ 17.89 GFLOPS | Progress: (12/20) | 9.46 s
[Task 5/25] Current/Best: 15.31/ 22.74 GFLOPS | Progress: (16/20) | 11.43 s
[Task 5/25] Current/Best: 14.06/ 22.74 GFLOPS | Progress: (20/20) | 13.23 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 18.51/ 18.51 GFLOPS | Progress: (4/20) | 5.73 s
[Task 6/25] Current/Best: 12.63/ 18.51 GFLOPS | Progress: (8/20) | 8.53 s
[Task 6/25] Current/Best: 2.68/ 18.51 GFLOPS | Progress: (12/20) | 14.67 s
[Task 6/25] Current/Best: 12.89/ 18.51 GFLOPS | Progress: (16/20) | 17.06 s
[Task 6/25] Current/Best: 9.67/ 18.51 GFLOPS | Progress: (20/20) | 19.89 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 5.12/ 11.94 GFLOPS | Progress: (4/20) | 5.91 s
[Task 7/25] Current/Best: 14.32/ 14.40 GFLOPS | Progress: (8/20) | 8.32 s
[Task 7/25] Current/Best: 13.88/ 19.83 GFLOPS | Progress: (12/20) | 10.56 s
[Task 7/25] Current/Best: 12.38/ 19.83 GFLOPS | Progress: (16/20) | 13.24 s
[Task 7/25] Current/Best: 10.86/ 19.83 GFLOPS | Progress: (20/20) | 16.73 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 14.25/ 14.25 GFLOPS | Progress: (4/20) | 5.88 s
[Task 8/25] Current/Best: 2.90/ 14.25 GFLOPS | Progress: (8/20) | 12.72 s
[Task 8/25] Current/Best: 12.96/ 17.72 GFLOPS | Progress: (12/20) | 15.58 s
[Task 8/25] Current/Best: 3.89/ 17.72 GFLOPS | Progress: (16/20) | 18.53 s
[Task 8/25] Current/Best: 11.17/ 17.72 GFLOPS | Progress: (20/20) | 21.12 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 13.34/ 13.39 GFLOPS | Progress: (4/20) | 4.79 s
[Task 9/25] Current/Best: 7.52/ 22.70 GFLOPS | Progress: (8/20) | 6.87 s
[Task 9/25] Current/Best: 12.80/ 22.70 GFLOPS | Progress: (12/20) | 14.59 s
[Task 9/25] Current/Best: 6.18/ 22.70 GFLOPS | Progress: (16/20) | 17.83 s
[Task 9/25] Current/Best: 6.71/ 22.70 GFLOPS | Progress: (20/20) | 25.11 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 14.29/ 16.65 GFLOPS | Progress: (4/20) | 5.36 s
[Task 10/25] Current/Best: 13.72/ 18.09 GFLOPS | Progress: (8/20) | 7.86 s
[Task 10/25] Current/Best: 13.41/ 18.09 GFLOPS | Progress: (12/20) | 9.79 s
[Task 10/25] Current/Best: 11.74/ 18.09 GFLOPS | Progress: (16/20) | 12.66 s
[Task 10/25] Current/Best: 14.05/ 18.09 GFLOPS | Progress: (20/20) | 15.24 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 17.89/ 18.16 GFLOPS | Progress: (4/20) | 6.49 s
[Task 11/25] Current/Best: 6.68/ 18.16 GFLOPS | Progress: (8/20) | 9.83 s
[Task 11/25] Current/Best: 12.14/ 18.16 GFLOPS | Progress: (12/20) | 12.14 s
[Task 11/25] Current/Best: 13.07/ 19.41 GFLOPS | Progress: (16/20) | 14.60 s
[Task 11/25] Current/Best: 7.01/ 19.41 GFLOPS | Progress: (20/20) | 18.85 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 12.72/ 15.29 GFLOPS | Progress: (4/20) | 5.69 s
[Task 12/25] Current/Best: 16.70/ 16.70 GFLOPS | Progress: (8/20) | 9.61 s
[Task 12/25] Current/Best: 15.00/ 16.70 GFLOPS | Progress: (12/20) | 12.52 s
[Task 12/25] Current/Best: 4.87/ 16.70 GFLOPS | Progress: (16/20) | 14.72 s
[Task 12/25] Current/Best: 13.23/ 16.70 GFLOPS | Progress: (20/20) | 17.20 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 6.27/ 17.05 GFLOPS | Progress: (4/20) | 5.75 s
[Task 13/25] Current/Best: 9.16/ 18.20 GFLOPS | Progress: (8/20) | 9.30 s
[Task 13/25] Current/Best: 18.34/ 21.25 GFLOPS | Progress: (12/20) | 11.59 s
[Task 13/25] Current/Best: 14.32/ 21.25 GFLOPS | Progress: (16/20) | 14.93 s
[Task 13/25] Current/Best: 7.79/ 21.25 GFLOPS | Progress: (20/20) | 18.60 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 9.99/ 19.37 GFLOPS | Progress: (4/20) | 5.16 s
[Task 14/25] Current/Best: 18.76/ 19.57 GFLOPS | Progress: (8/20) | 8.33 s
[Task 14/25] Current/Best: 14.21/ 19.57 GFLOPS | Progress: (12/20) | 10.73 s
[Task 14/25] Current/Best: 10.99/ 19.57 GFLOPS | Progress: (16/20) | 15.09 s
[Task 14/25] Current/Best: 7.45/ 20.77 GFLOPS | Progress: (20/20) | 21.20 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 18.34/ 18.58 GFLOPS | Progress: (4/20) | 9.11 s
[Task 15/25] Current/Best: 18.24/ 18.58 GFLOPS | Progress: (8/20) | 11.32 s
[Task 15/25] Current/Best: 10.08/ 18.62 GFLOPS | Progress: (12/20) | 16.11 s
[Task 15/25] Current/Best: 9.78/ 23.75 GFLOPS | Progress: (16/20) | 17.87 s Done.
-
[Task 15/25] Current/Best: 12.43/ 23.75 GFLOPS | Progress: (20/20) | 21.19 s Done.
-
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 16.90/ 16.94 GFLOPS | Progress: (4/20) | 4.88 s
[Task 16/25] Current/Best: 11.53/ 19.19 GFLOPS | Progress: (8/20) | 6.43 s
[Task 16/25] Current/Best: 16.22/ 19.19 GFLOPS | Progress: (12/20) | 8.85 s
[Task 16/25] Current/Best: 15.46/ 19.19 GFLOPS | Progress: (16/20) | 11.36 s
[Task 16/25] Current/Best: 16.06/ 19.19 GFLOPS | Progress: (20/20) | 14.91 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 13.35/ 20.21 GFLOPS | Progress: (4/20) | 5.78 s
[Task 17/25] Current/Best: 19.57/ 20.21 GFLOPS | Progress: (8/20) | 8.19 s
[Task 17/25] Current/Best: 9.60/ 20.21 GFLOPS | Progress: (12/20) | 10.52 s
[Task 17/25] Current/Best: 16.39/ 20.21 GFLOPS | Progress: (16/20) | 14.14 s
[Task 17/25] Current/Best: 17.27/ 20.21 GFLOPS | Progress: (20/20) | 16.92 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 16.36/ 18.95 GFLOPS | Progress: (4/20) | 6.80 s
[Task 18/25] Current/Best: 8.35/ 18.95 GFLOPS | Progress: (8/20) | 11.24 s
[Task 18/25] Current/Best: 10.75/ 18.95 GFLOPS | Progress: (12/20) | 18.52 s
[Task 18/25] Current/Best: 10.53/ 18.95 GFLOPS | Progress: (16/20) | 23.20 s
[Task 18/25] Current/Best: 17.73/ 18.95 GFLOPS | Progress: (20/20) | 25.38 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 11.84/ 11.84 GFLOPS | Progress: (4/20) | 5.85 s
[Task 19/25] Current/Best: 17.88/ 19.34 GFLOPS | Progress: (8/20) | 8.55 s
[Task 19/25] Current/Best: 5.22/ 19.34 GFLOPS | Progress: (12/20) | 13.99 s
[Task 19/25] Current/Best: 21.50/ 21.50 GFLOPS | Progress: (16/20) | 16.74 s
[Task 19/25] Current/Best: 18.54/ 21.50 GFLOPS | Progress: (20/20) | 19.47 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 11.39/ 15.39 GFLOPS | Progress: (4/20) | 5.79 s
[Task 20/25] Current/Best: 12.86/ 15.39 GFLOPS | Progress: (8/20) | 8.34 s
[Task 20/25] Current/Best: 9.24/ 17.52 GFLOPS | Progress: (12/20) | 12.10 s
[Task 20/25] Current/Best: 16.90/ 18.40 GFLOPS | Progress: (16/20) | 14.24 s
[Task 20/25] Current/Best: 14.35/ 18.40 GFLOPS | Progress: (20/20) | 18.49 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 1.61/ 16.19 GFLOPS | Progress: (4/20) | 5.29 s
[Task 21/25] Current/Best: 20.65/ 20.65 GFLOPS | Progress: (8/20) | 8.23 s
[Task 21/25] Current/Best: 18.79/ 20.65 GFLOPS | Progress: (12/20) | 10.31 s Done.
-
[Task 21/25] Current/Best: 2.71/ 20.65 GFLOPS | Progress: (16/20) | 14.13 s
[Task 21/25] Current/Best: 5.32/ 20.65 GFLOPS | Progress: (20/20) | 17.80 s Done.
-
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 7.19/ 16.01 GFLOPS | Progress: (4/20) | 5.43 s
[Task 22/25] Current/Best: 13.58/ 20.96 GFLOPS | Progress: (8/20) | 7.64 s
[Task 22/25] Current/Best: 6.82/ 20.96 GFLOPS | Progress: (12/20) | 9.81 s
[Task 22/25] Current/Best: 3.08/ 20.96 GFLOPS | Progress: (16/20) | 11.87 s
[Task 22/25] Current/Best: 14.79/ 20.96 GFLOPS | Progress: (20/20) | 13.55 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 11.49/ 19.69 GFLOPS | Progress: (4/20) | 5.91 s
[Task 23/25] Current/Best: 18.36/ 19.69 GFLOPS | Progress: (8/20) | 9.95 s
[Task 23/25] Current/Best: 5.36/ 19.69 GFLOPS | Progress: (12/20) | 14.29 s
[Task 23/25] Current/Best: 19.17/ 19.69 GFLOPS | Progress: (16/20) | 17.24 s
[Task 23/25] Current/Best: 21.44/ 22.08 GFLOPS | Progress: (20/20) | 19.33 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 7.30/ 7.30 GFLOPS | Progress: (4/20) | 13.87 s
[Task 24/25] Current/Best: 2.12/ 7.30 GFLOPS | Progress: (8/20) | 24.94 s
[Task 24/25] Current/Best: 8.28/ 8.28 GFLOPS | Progress: (12/20) | 36.94 s
[Task 24/25] Current/Best: 2.25/ 8.28 GFLOPS | Progress: (16/20) | 47.96 s
[Task 24/25] Current/Best: 6.86/ 9.16 GFLOPS | Progress: (20/20) | 50.78 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 25/25] Current/Best: 5.27/ 8.26 GFLOPS | Progress: (4/20) | 4.28 s
[Task 25/25] Current/Best: 8.17/ 8.26 GFLOPS | Progress: (8/20) | 9.43 s
[Task 25/25] Current/Best: 8.48/ 8.77 GFLOPS | Progress: (12/20) | 11.10 s
[Task 25/25] Current/Best: 1.53/ 8.77 GFLOPS | Progress: (16/20) | 21.79 s
[Task 25/25] Current/Best: 4.13/ 8.78 GFLOPS | Progress: (20/20) | 34.29 s
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 6.53/ 12.24 GFLOPS | Progress: (4/20) | 13.43 s
[Task 1/25] Current/Best: 12.81/ 23.55 GFLOPS | Progress: (8/20) | 17.21 s
[Task 1/25] Current/Best: 11.15/ 23.55 GFLOPS | Progress: (12/20) | 20.21 s
[Task 1/25] Current/Best: 16.32/ 23.55 GFLOPS | Progress: (16/20) | 22.55 s
[Task 1/25] Current/Best: 11.62/ 23.55 GFLOPS | Progress: (20/20) | 25.91 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 5.87/ 16.30 GFLOPS | Progress: (4/20) | 4.90 s
[Task 2/25] Current/Best: 17.02/ 17.76 GFLOPS | Progress: (8/20) | 6.90 s
[Task 2/25] Current/Best: 10.99/ 17.76 GFLOPS | Progress: (12/20) | 10.08 s
[Task 2/25] Current/Best: 17.00/ 17.76 GFLOPS | Progress: (16/20) | 11.96 s
[Task 2/25] Current/Best: 12.60/ 19.37 GFLOPS | Progress: (20/20) | 14.03 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 18.62/ 18.62 GFLOPS | Progress: (4/20) | 5.63 s
[Task 3/25] Current/Best: 10.14/ 18.62 GFLOPS | Progress: (8/20) | 8.54 s
[Task 3/25] Current/Best: 5.72/ 21.38 GFLOPS | Progress: (12/20) | 10.99 s
[Task 3/25] Current/Best: 15.90/ 21.38 GFLOPS | Progress: (16/20) | 14.80 s
[Task 3/25] Current/Best: 22.62/ 22.62 GFLOPS | Progress: (20/20) | 17.41 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 7.90/ 16.33 GFLOPS | Progress: (4/20) | 6.07 s
[Task 4/25] Current/Best: 16.96/ 16.96 GFLOPS | Progress: (8/20) | 7.91 s
[Task 4/25] Current/Best: 12.64/ 16.96 GFLOPS | Progress: (12/20) | 10.37 s
[Task 4/25] Current/Best: 10.92/ 16.96 GFLOPS | Progress: (16/20) | 13.46 s
[Task 4/25] Current/Best: 5.57/ 17.01 GFLOPS | Progress: (20/20) | 16.49 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 13.90/ 15.72 GFLOPS | Progress: (4/20) | 5.11 s
[Task 5/25] Current/Best: 11.91/ 15.72 GFLOPS | Progress: (8/20) | 9.36 s
[Task 5/25] Current/Best: 19.53/ 19.53 GFLOPS | Progress: (12/20) | 11.97 s
[Task 5/25] Current/Best: 11.55/ 20.31 GFLOPS | Progress: (16/20) | 14.46 s
[Task 5/25] Current/Best: 22.59/ 22.59 GFLOPS | Progress: (20/20) | 16.04 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 17.18/ 17.18 GFLOPS | Progress: (4/20) | 5.29 s
[Task 6/25] Current/Best: 12.67/ 19.02 GFLOPS | Progress: (8/20) | 7.62 s
[Task 6/25] Current/Best: 9.09/ 19.02 GFLOPS | Progress: (12/20) | 10.65 s
[Task 6/25] Current/Best: 10.72/ 19.02 GFLOPS | Progress: (16/20) | 14.55 s
[Task 6/25] Current/Best: 14.11/ 19.02 GFLOPS | Progress: (20/20) | 17.29 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 5.42/ 19.41 GFLOPS | Progress: (4/20) | 5.49 s
[Task 7/25] Current/Best: 6.16/ 19.41 GFLOPS | Progress: (8/20) | 8.11 s
[Task 7/25] Current/Best: 11.41/ 19.41 GFLOPS | Progress: (12/20) | 11.21 s
[Task 7/25] Current/Best: 5.62/ 22.24 GFLOPS | Progress: (16/20) | 13.52 s
[Task 7/25] Current/Best: 9.53/ 22.48 GFLOPS | Progress: (20/20) | 15.82 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 7.72/ 13.39 GFLOPS | Progress: (4/20) | 5.34 s
[Task 8/25] Current/Best: 10.23/ 18.16 GFLOPS | Progress: (8/20) | 17.25 s
[Task 8/25] Current/Best: 9.91/ 18.16 GFLOPS | Progress: (12/20) | 23.93 s
[Task 8/25] Current/Best: 15.29/ 18.16 GFLOPS | Progress: (16/20) | 32.91 s
[Task 8/25] Current/Best: 8.35/ 18.16 GFLOPS | Progress: (20/20) | 37.83 s
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 20.80/ 20.80 GFLOPS | Progress: (4/20) | 5.19 s Done.
+
[Task 9/25] Current/Best: 8.93/ 20.80 GFLOPS | Progress: (8/20) | 13.28 s
[Task 9/25] Current/Best: 7.85/ 20.80 GFLOPS | Progress: (12/20) | 15.15 s
[Task 9/25] Current/Best: 16.72/ 20.80 GFLOPS | Progress: (16/20) | 17.25 s
[Task 9/25] Current/Best: 17.64/ 20.80 GFLOPS | Progress: (20/20) | 18.91 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 6.19/ 10.90 GFLOPS | Progress: (4/20) | 5.45 s
[Task 10/25] Current/Best: 3.13/ 11.28 GFLOPS | Progress: (8/20) | 8.67 s
[Task 10/25] Current/Best: 5.03/ 17.04 GFLOPS | Progress: (12/20) | 11.07 s
[Task 10/25] Current/Best: 17.39/ 17.39 GFLOPS | Progress: (16/20) | 13.23 s
[Task 10/25] Current/Best: 18.30/ 19.61 GFLOPS | Progress: (20/20) | 14.96 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 13.24/ 13.24 GFLOPS | Progress: (4/20) | 6.14 s
[Task 11/25] Current/Best: 14.24/ 18.26 GFLOPS | Progress: (8/20) | 9.17 s
[Task 11/25] Current/Best: 11.59/ 18.94 GFLOPS | Progress: (12/20) | 11.82 s
[Task 11/25] Current/Best: 3.08/ 19.08 GFLOPS | Progress: (16/20) | 14.53 s
[Task 11/25] Current/Best: 9.39/ 21.22 GFLOPS | Progress: (20/20) | 17.31 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 11.30/ 12.37 GFLOPS | Progress: (4/20) | 8.39 s
[Task 12/25] Current/Best: 18.24/ 18.24 GFLOPS | Progress: (8/20) | 10.83 s
[Task 12/25] Current/Best: 10.24/ 18.24 GFLOPS | Progress: (12/20) | 20.60 s
[Task 12/25] Current/Best: 9.85/ 18.24 GFLOPS | Progress: (16/20) | 25.59 s
[Task 12/25] Current/Best: 3.87/ 18.24 GFLOPS | Progress: (20/20) | 28.76 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 16.50/ 16.50 GFLOPS | Progress: (4/20) | 6.95 s
[Task 13/25] Current/Best: 7.56/ 19.99 GFLOPS | Progress: (8/20) | 9.88 s
[Task 13/25] Current/Best: 9.64/ 19.99 GFLOPS | Progress: (12/20) | 14.11 s
[Task 13/25] Current/Best: 9.73/ 20.75 GFLOPS | Progress: (16/20) | 17.60 s
[Task 13/25] Current/Best: 13.20/ 20.75 GFLOPS | Progress: (20/20) | 20.55 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 14.97/ 21.77 GFLOPS | Progress: (4/20) | 6.72 s
[Task 14/25] Current/Best: 8.84/ 21.77 GFLOPS | Progress: (8/20) | 9.03 s
[Task 14/25] Current/Best: 10.00/ 21.77 GFLOPS | Progress: (12/20) | 13.21 s
[Task 14/25] Current/Best: 7.80/ 21.77 GFLOPS | Progress: (16/20) | 17.34 s
[Task 14/25] Current/Best: 13.66/ 21.77 GFLOPS | Progress: (20/20) | 19.59 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 17.51/ 19.96 GFLOPS | Progress: (4/20) | 5.87 s
[Task 15/25] Current/Best: 20.70/ 20.70 GFLOPS | Progress: (8/20) | 7.98 s
[Task 15/25] Current/Best: 6.56/ 20.70 GFLOPS | Progress: (12/20) | 9.77 s
[Task 15/25] Current/Best: 20.54/ 20.70 GFLOPS | Progress: (16/20) | 14.74 s
[Task 15/25] Current/Best: 4.91/ 20.70 GFLOPS | Progress: (20/20)
| 17.76 s Done.
+
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 13.47/ 14.65 GFLOPS | Progress: (4/20) | 5.45 s
[Task 16/25] Current/Best: 4.40/ 15.73 GFLOPS | Progress: (8/20) | 7.48 s
[Task 16/25] Current/Best: 15.40/ 19.15 GFLOPS | Progress: (12/20) | 9.29 s
[Task 16/25] Current/Best: 15.45/ 19.15 GFLOPS | Progress: (16/20) | 11.08 s
[Task 16/25] Current/Best: 19.51/ 19.51 GFLOPS | Progress: (20/20) | 13.41 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 6.19/ 14.96 GFLOPS | Progress: (4/20) | 6.31 s
[Task 17/25] Current/Best: 6.19/ 17.22 GFLOPS | Progress: (8/20) | 9.37 s
[Task 17/25] Current/Best: 13.17/ 21.57 GFLOPS | Progress: (12/20) | 11.77 s
[Task 17/25] Current/Best: 17.74/ 21.57 GFLOPS | Progress: (16/20) | 15.47 s
[Task 17/25] Current/Best: 18.20/ 21.64 GFLOPS | Progress: (20/20) | 17.94 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 16.46/ 20.79 GFLOPS | Progress: (4/20) | 5.22 s
[Task 18/25] Current/Best: 18.04/ 20.79 GFLOPS | Progress: (8/20) | 7.82 s
[Task 18/25] Current/Best: 12.73/ 20.79 GFLOPS | Progress: (12/20) | 10.09 s
[Task 18/25] Current/Best: 8.54/ 20.79 GFLOPS | Progress: (16/20) | 17.65 s
[Task 18/25] Current/Best: 8.35/ 20.79 GFLOPS | Progress: (20/20) | 25.71 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 6.10/ 10.75 GFLOPS | Progress: (4/20) | 7.01 s
[Task 19/25] Current/Best: 5.38/ 10.75 GFLOPS | Progress: (8/20) | 11.64 s
[Task 19/25] Current/Best: 11.18/ 11.18 GFLOPS | Progress: (12/20) | 15.24 s
[Task 19/25] Current/Best: 7.90/ 12.33 GFLOPS | Progress: (16/20) | 19.12 s
[Task 19/25] Current/Best: 12.87/ 12.87 GFLOPS | Progress: (20/20) | 23.24 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 4.79/ 15.94 GFLOPS | Progress: (4/20) | 6.09 s
[Task 20/25] Current/Best: 10.51/ 15.94 GFLOPS | Progress: (8/20) | 7.91 s
[Task 20/25] Current/Best: 2.17/ 15.94 GFLOPS | Progress: (12/20) | 13.69 s
[Task 20/25] Current/Best: 18.47/ 18.47 GFLOPS | Progress: (16/20) | 16.67 s
[Task 20/25] Current/Best: 15.48/ 18.47 GFLOPS | Progress: (20/20) | 18.99 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+ Done.
+
[Task 21/25] Current/Best: 15.27/ 18.40 GFLOPS | Progress: (4/20) | 6.21 s
[Task 21/25] Current/Best: 16.05/ 18.40 GFLOPS | Progress: (8/20) | 9.82 s
[Task 21/25] Current/Best: 4.67/ 18.40 GFLOPS | Progress: (12/20) | 12.02 s
[Task 21/25] Current/Best: 15.27/ 18.40 GFLOPS | Progress: (16/20) | 15.61 s
[Task 21/25] Current/Best: 8.30/ 18.40 GFLOPS | Progress: (20/20) | 23.50 s Done.
+
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 9.64/ 9.64 GFLOPS | Progress: (4/20) | 6.69 s
[Task 22/25] Current/Best: 6.12/ 16.95 GFLOPS | Progress: (8/20) | 9.23 s
[Task 22/25] Current/Best: 5.00/ 16.95 GFLOPS | Progress: (12/20) | 11.34 s
[Task 22/25] Current/Best: 13.53/ 16.95 GFLOPS | Progress: (16/20) | 13.76 s
[Task 22/25] Current/Best: 4.71/ 17.39 GFLOPS | Progress: (20/20) | 17.25 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.88/ 17.88 GFLOPS | Progress: (4/20) | 6.72 s
[Task 23/25] Current/Best: 18.12/ 18.12 GFLOPS | Progress: (8/20) | 9.96 s
[Task 23/25] Current/Best: 20.81/ 21.78 GFLOPS | Progress: (12/20) | 13.48 s
[Task 23/25] Current/Best: 8.77/ 21.78 GFLOPS | Progress: (16/20) | 19.58 s
[Task 23/25] Current/Best: 17.72/ 21.78 GFLOPS | Progress: (20/20) | 22.68 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 6.85/ 6.85 GFLOPS | Progress: (4/20) | 13.76 s
[Task 24/25] Current/Best: 1.66/ 6.85 GFLOPS | Progress: (8/20) | 16.98 s
[Task 24/25] Current/Best: 1.93/ 6.85 GFLOPS | Progress: (12/20) | 28.50 s
[Task 24/25] Current/Best: 5.41/ 6.85 GFLOPS | Progress: (16/20) | 39.54 s
[Task 24/25] Current/Best: 3.04/ 6.85 GFLOPS | Progress: (20/20) | 50.59 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 5.54/ 5.54 GFLOPS | Progress: (4/20) | 14.03 s Done.
+
[Task 25/25] Current/Best: 3.53/ 8.25 GFLOPS | Progress: (8/20) | 17.30 s
[Task 25/25] Current/Best: 8.17/ 8.25 GFLOPS | Progress: (12/20) | 28.31 s
[Task 25/25] Current/Best: 6.74/ 8.25 GFLOPS | Progress: (16/20) | 31.74 s
[Task 25/25] Current/Best: 4.39/ 8.25 GFLOPS | Progress: (20/20) | 33.45 s
@@ -702,8 +702,8 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.621102
- class='n02123159 tiger cat' with probability=0.356380
+ class='n02123045 tabby, tabby cat' with probability=0.621103
+ class='n02123159 tiger cat' with probability=0.356379
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
class='n04040759 radiator' with probability=0.000262
@@ -760,8 +760,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 425.3929725300077, 'median': 423.1916936000289, 'std': 4.688961294016269}
- unoptimized: {'mean': 513.6553296399961, 'median': 513.3987593500024, 'std': 1.727023582978912}
+ optimized: {'mean': 436.6873022300001, 'median': 435.1963521500011, 'std': 4.867030740095508}
+ unoptimized: {'mean': 524.1644680799982, 'median': 524.1663549999998, 'std': 0.5732194644729002}
@@ -784,7 +784,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 12 minutes 47.002 seconds)
+ **Total running time of the script:** ( 13 minutes 13.634 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 fcd74f9327..23a3fd005e 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
- 1.244e-07 secs/op
+ 1.229e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 18bbec7338..6e990b1b85 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -270,7 +270,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x25602320)), stage(b, placeholder(b, 0x290194c0)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T [...]
+ [stage(a, placeholder(a, 0x7e848a0)), stage(b, placeholder(b, 0x7eb9bb0)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.R [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index c3e94baff1..8974528e14 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,29 +5,29 @@
Computation times
=================
-**16:32.604** total execution time for **tutorial** files:
+**17:07.701** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 12:47.002 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 13:13.634 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:28.212 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:37.901 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.411 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:04.804 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:38.222 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:39.455 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:35.959 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:28.984 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.870 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.813 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.725 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.898 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.203 | 0.0 MB |
-+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.000 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.210 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.000 | 0.0 MB |
++------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.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 4ad93c212d..761d919ee9 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: 0.000009
naive: 0.000007
@@ -444,7 +444,7 @@ factor to be the number of threads on your CPU.
.. code-block:: none
- vector: 0.000040
+ vector: 0.000042
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -498,10 +498,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 7.417429997076397e-06 1.0
- naive 6.8533999999999995e-06 0.9239588378591086
- parallel 7.3477e-06 0.9905991701837589
- vector 4.04322e-05 5.4509715650752995
+ numpy 8.515459999216546e-06 1.0
+ naive 6.815899999999999e-06 0.8004147750828596
+ parallel 7.0470000000000005e-06 0.8275536495560253
+ vector 4.2298999999999996e-05 4.967318266293502
@@ -922,7 +922,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.019570
+ Numpy running time: 0.020702
@@ -980,7 +980,7 @@ optimizations.
.. code-block:: none
- none: 3.400834
+ none: 3.593863
@@ -1080,7 +1080,7 @@ schedule.
.. code-block:: none
- blocking: 0.315847
+ blocking: 0.333657
@@ -1164,7 +1164,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.344345
+ vectorization: 0.353971
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1230,7 +1230,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.121476
+ loop permutation: 0.145312
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1321,7 +1321,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.108671
+ array packing: 0.112158
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1404,7 +1404,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110471
+ block caching: 0.114877
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1478,7 +1478,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.146984
+ parallelization: 0.148319
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1548,13 +1548,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.4008339759000004 1.0
- blocking 0.3158472607 0.09287347248888086
- vectorization 0.3443454349 0.10125323298349843
- loop permutation 0.12147584770000001 0.035719428987371404
- array packing 0.10867131349999999 0.03195431305088661
- block caching 0.1104708708 0.03248346481564566
- parallelization 0.14698438490000001 0.04322010011120931
+ none 3.593863143 1.0
+ blocking 0.333657166 0.09284081021557142
+ vectorization 0.3539705903 0.0984930633737268
+ loop permutation 0.1453124582 0.04043349799867434
+ array packing 0.11215777869999999 0.031208138495328337
+ block caching 0.11487654150000001 0.03196463997905777
+ parallelization 0.14831938130000002 0.04127018069368926
@@ -1596,7 +1596,7 @@ the computation for specific platforms.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.411 seconds)
+ **Total running time of the script:** ( 1 minutes 4.804 seconds)
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 67f9068e1f..83e740dd21 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-f4b53fb6e259049a03d4ab8099013c4b3270c6ae
+af17cfeb4c71fdfb7f062238859abdf1479fa48f
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index aee943c07c..44fe1344ad 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -590,7 +590,7 @@ class:['truck 0.9266'] left:471 top:83 right:689 bottom:169
class:['bicycle 0.9984'] 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 25.899 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 27.632 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-darknet-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7716f96385bd5abb6e822041e285be54/from_darknet.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_darknet.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 3d3f1216e0..3884a19887 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -444,7 +444,7 @@
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</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.zipc1772097-6190-47d0-863c-9dddc2a5c20e 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.zip8b605d40-69c4-4d1a-b2c5-692bee6cd627 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 fbcaa82dab..49747d92a4 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -454,14 +454,13 @@ Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdo
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "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, 63.1MB/s]
- 30%|##9 | 12.3M/41.5M [00:00<00:00, 47.0MB/s]
- 41%|####1 | 17.0M/41.5M [00:00<00:00, 45.0MB/s]
- 54%|#####3 | 22.3M/41.5M [00:00<00:00, 47.4MB/s]
- 65%|######4 | 26.9M/41.5M [00:00<00:00, 41.9MB/s]
- 77%|#######7 | 32.0M/41.5M [00:00<00:00, 43.3MB/s]
- 92%|#########2| 38.3M/41.5M [00:01<00:00, 32.7MB/s]
-100%|##########| 41.5M/41.5M [00:01<00:00, 37.2MB/s]
+ 15%|#5 | 6.33M/41.5M [00:00<00:00, 38.5MB/s]
+ 24%|##4 | 10.0M/41.5M [00:00<00:01, 31.0MB/s]
+ 39%|###8 | 16.0M/41.5M [00:00<00:00, 34.8MB/s]
+ 58%|#####7 | 24.0M/41.5M [00:00<00:00, 45.1MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 48.1MB/s]
+ 92%|#########2| 38.3M/41.5M [00:00<00:00, 43.8MB/s]
+100%|##########| 41.5M/41.5M [00:01<00:00, 42.9MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_paddle.html b/docs/how_to/compile_models/from_paddle.html
index cd86dc8824..8d106e2943 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -489,6 +489,7 @@ To begin, we’ll install PaddlePaddle>=2.1.3:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>TVM prediction top-1 id: 282, class name: 282: 'tiger cat',
</pre></div>
</div>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 2.603 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-paddle-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/16269b77359771348d507395692524cf/from_paddle.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_paddle.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index b9d12cbd70..52425f6b62 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -437,13 +437,12 @@ be unstable.</p>
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, 70.2MB/s]
- 36%|###5 | 16.0M/44.7M [00:00<00:00, 55.9MB/s]
- 54%|#####3 | 24.0M/44.7M [00:00<00:00, 61.2MB/s]
- 72%|#######1 | 32.0M/44.7M [00:00<00:00, 60.3MB/s]
- 86%|########5 | 38.3M/44.7M [00:00<00:00, 35.5MB/s]
- 96%|#########5| 42.7M/44.7M [00:01<00:00, 32.4MB/s]
-100%|##########| 44.7M/44.7M [00:01<00:00, 42.0MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 53.0MB/s]
+ 32%|###2 | 14.3M/44.7M [00:00<00:00, 55.1MB/s]
+ 49%|####9 | 22.1M/44.7M [00:00<00:00, 65.4MB/s]
+ 69%|######9 | 30.9M/44.7M [00:00<00:00, 75.1MB/s]
+ 86%|########5 | 38.3M/44.7M [00:00<00:00, 72.1MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 63.2MB/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 9be0b19e5b..430905fbaf 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -657,7 +657,7 @@ banana (score = 0.00022)
desk (score = 0.00019)
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 30.395 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 36.993 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 85ea312e3d..abd8faf420 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>07:01.986</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>07:22.248</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -354,43 +354,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></td>
-<td><p>01:30.395</p></td>
+<td><p>01:36.993</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>01:25.899</p></td>
+<td><p>01:27.632</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>00:58.807</p></td>
+<td><p>01:02.603</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></td>
-<td><p>00:38.578</p></td>
+<td><p>00:40.561</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
-<td><p>00:33.808</p></td>
+<td><p>00:35.433</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></td>
-<td><p>00:31.132</p></td>
+<td><p>00:32.282</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>00:28.431</p></td>
+<td><p>00:29.077</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></td>
-<td><p>00:26.973</p></td>
+<td><p>00:28.335</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>00:25.132</p></td>
+<td><p>00:26.464</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>00:02.831</p></td>
+<td><p>00:02.868</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 2bae44d922..bc9921b43a 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -835,7 +835,7 @@ Top5 predictions:
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 2752.7610 2752.9445 2755.0147 2751.1536 1.2200
+ 2754.4285 2753.8857 2758.8902 2751.5141 2.0346
</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_adreno_tvmc.html b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
index 4941230b90..9f4988104e 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
@@ -444,19 +444,17 @@ to run this tutorial with a real device over rpc.</p>
8192/102967424 [..............................] - ETA: 0s
8380416/102967424 [=>............................] - ETA: 1s
- 15024128/102967424 [===>..........................] - ETA: 0s
+ 15024128/102967424 [===>..........................] - ETA: 1s
16769024/102967424 [===>..........................] - ETA: 1s
- 19611648/102967424 [====>.........................] - ETA: 1s
- 22970368/102967424 [=====>........................] - ETA: 1s
- 23412736/102967424 [=====>........................] - ETA: 1s
+ 24027136/102967424 [======>.......................] - ETA: 1s
25157632/102967424 [======>.......................] - ETA: 1s
- 31498240/102967424 [========>.....................] - ETA: 1s
+ 31899648/102967424 [========>.....................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
- 40189952/102967424 [==========>...................] - ETA: 1s
41934848/102967424 [===========>..................] - ETA: 1s
- 48578560/102967424 [=============>................] - ETA: 1s
+ 50135040/102967424 [=============>................] - ETA: 0s
50323456/102967424 [=============>................] - ETA: 1s
- 58712064/102967424 [================>.............] - ETA: 1s
+ 56967168/102967424 [===============>..............] - ETA: 0s
+ 58712064/102967424 [================>.............] - ETA: 0s
65355776/102967424 [==================>...........] - ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] - ETA: 0s
@@ -465,7 +463,7 @@ to run this tutorial with a real device over rpc.</p>
83877888/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
- 98910208/102967424 [===========================>..] - ETA: 0s
+100646912/102967424 [============================>.] - ETA: 0s
102850560/102967424 [============================>.] - ETA: 0s
102967424/102967424 [==============================] - 2s 0us/step
</pre></div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 499e96d277..24bd347870 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -667,7 +667,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.3212 16.2892 16.4534 16.2165 0.0857
+ 16.8435 16.8438 17.2434 16.6309 0.1515
</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 6abcbf38aa..3883524f85 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -459,32 +459,33 @@ be unstable.</p>
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]
- 4%|3 | 6.30M/170M [00:00<00:03, 44.0MB/s]
- 6%|6 | 10.5M/170M [00:00<00:03, 43.1MB/s]
- 9%|8 | 14.6M/170M [00:00<00:05, 29.8MB/s]
- 10%|# | 17.7M/170M [00:00<00:05, 29.1MB/s]
- 14%|#4 | 24.0M/170M [00:00<00:04, 33.8MB/s]
- 19%|#8 | 32.0M/170M [00:00<00:03, 39.2MB/s]
- 24%|##3 | 40.0M/170M [00:01<00:03, 42.8MB/s]
- 28%|##8 | 48.0M/170M [00:01<00:02, 43.5MB/s]
- 31%|### | 52.1M/170M [00:01<00:03, 36.7MB/s]
- 33%|###2 | 56.0M/170M [00:01<00:03, 36.7MB/s]
- 38%|###7 | 64.0M/170M [00:01<00:03, 36.6MB/s]
- 42%|####2 | 72.0M/170M [00:01<00:02, 42.4MB/s]
- 47%|####7 | 80.0M/170M [00:02<00:02, 46.9MB/s]
- 52%|#####1 | 88.0M/170M [00:02<00:01, 48.7MB/s]
- 56%|#####5 | 94.3M/170M [00:02<00:01, 52.2MB/s]
- 59%|#####8 | 99.5M/170M [00:02<00:01, 52.3MB/s]
- 62%|######1 | 105M/170M [00:02<00:01, 45.6MB/s]
- 66%|######5 | 112M/170M [00:02<00:01, 49.8MB/s]
- 71%|####### | 120M/170M [00:02<00:00, 57.4MB/s]
- 75%|#######5 | 128M/170M [00:02<00:00, 56.4MB/s]
- 80%|######## | 136M/170M [00:03<00:00, 52.9MB/s]
- 85%|########4 | 144M/170M [00:03<00:00, 51.0MB/s]
- 89%|########9 | 152M/170M [00:03<00:00, 45.0MB/s]
- 94%|#########4| 160M/170M [00:03<00:00, 46.1MB/s]
- 98%|#########7| 166M/170M [00:03<00:00, 48.4MB/s]
-100%|##########| 170M/170M [00:03<00:00, 45.1MB/s]
+ 5%|4 | 7.99M/170M [00:00<00:02, 59.8MB/s]
+ 8%|8 | 14.3M/170M [00:00<00:03, 48.0MB/s]
+ 11%|#1 | 19.0M/170M [00:00<00:03, 43.0MB/s]
+ 14%|#3 | 23.1M/170M [00:00<00:04, 38.4MB/s]
+ 16%|#5 | 26.7M/170M [00:00<00:04, 36.7MB/s]
+ 19%|#8 | 32.0M/170M [00:00<00:04, 33.2MB/s]
+ 24%|##3 | 40.0M/170M [00:01<00:03, 42.3MB/s]
+ 27%|##7 | 46.3M/170M [00:01<00:02, 46.8MB/s]
+ 30%|### | 51.0M/170M [00:01<00:03, 39.2MB/s]
+ 33%|###2 | 56.0M/170M [00:01<00:03, 38.5MB/s]
+ 37%|###6 | 62.3M/170M [00:01<00:02, 43.6MB/s]
+ 40%|#### | 68.7M/170M [00:01<00:02, 49.2MB/s]
+ 43%|####3 | 73.7M/170M [00:01<00:02, 46.4MB/s]
+ 47%|####7 | 80.0M/170M [00:01<00:02, 41.6MB/s]
+ 52%|#####1 | 88.0M/170M [00:02<00:02, 40.1MB/s]
+ 56%|#####5 | 94.3M/170M [00:02<00:01, 45.4MB/s]
+ 60%|#####9 | 101M/170M [00:02<00:01, 51.8MB/s]
+ 63%|######2 | 107M/170M [00:02<00:01, 36.5MB/s]
+ 66%|######5 | 112M/170M [00:02<00:01, 38.2MB/s]
+ 71%|####### | 120M/170M [00:02<00:01, 46.6MB/s]
+ 75%|#######5 | 128M/170M [00:03<00:00, 47.3MB/s]
+ 80%|######## | 136M/170M [00:03<00:00, 49.6MB/s]
+ 87%|########6 | 148M/170M [00:03<00:00, 65.5MB/s]
+ 91%|#########1| 155M/170M [00:03<00:00, 54.8MB/s]
+ 95%|#########4| 161M/170M [00:03<00:00, 50.5MB/s]
+ 98%|#########7| 166M/170M [00:03<00:00, 51.5MB/s]
+100%|##########| 170M/170M [00:03<00:00, 45.8MB/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=& [...]
@@ -582,7 +583,7 @@ 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 49.647 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 4.015 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 88e8e99feb..ae53b4de60 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -501,8 +501,7 @@ Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 39.5MB/s]
- 90%|########9 | 12.2M/13.6M [00:00<00:00, 28.5MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 33.1MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 46.5MB/s]
</pre></div>
</div>
</div>
@@ -593,7 +592,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.2833 90.2104 91.8116 90.0660 0.2546
+ 90.7178 90.5539 103.0548 90.2578 1.2598
</pre></div>
</div>
<div class="admonition note">
@@ -632,7 +631,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
<div class="section" id="deploy-a-quantized-tflite-model">
<h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
<p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.763 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.693 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 dffb4cbade..1f0f09829e 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -585,7 +585,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 119.9694 119.9506 121.3017 119.1650 0.3862
+ 122.4928 122.4673 125.0076 121.6657 0.3962
</pre></div>
</div>
<div class="admonition note">
@@ -613,7 +613,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 31.029 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 37.759 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 a1a755318e..463d1a8bde 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -526,7 +526,7 @@ for calibration. But the accuracy might be impacted.</p>
DeprecationWarning,
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 53.156 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 49.957 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 8259dd69da..03275948f3 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -468,24 +468,23 @@ 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<?, ?KB/s]
- 4%|4 | 5596/132723 [00:00<00:02, 55955.38KB/s]
- 10%|9 | 13038/132723 [00:00<00:01, 66812.84KB/s]
- 15%|#5 | 20544/132723 [00:00<00:01, 70575.90KB/s]
- 21%|##1 | 28379/132723 [00:00<00:01, 73642.54KB/s]
- 27%|##7 | 36257/132723 [00:00<00:01, 75493.91KB/s]
- 33%|###3 | 44136/132723 [00:00<00:01, 76612.75KB/s]
- 39%|###9 | 52063/132723 [00:00<00:01, 77479.88KB/s]
- 45%|####5 | 59974/132723 [00:00<00:00, 77995.99KB/s]
- 51%|#####1 | 67774/132723 [00:01<00:01, 58530.94KB/s]
- 57%|#####6 | 75637/132723 [00:01<00:00, 63573.63KB/s]
- 62%|######2 | 82588/132723 [00:01<00:00, 55777.49KB/s]
- 68%|######7 | 89993/132723 [00:01<00:00, 60235.43KB/s]
- 73%|#######3 | 97548/132723 [00:01<00:00, 64192.59KB/s]
- 79%|#######8 | 104377/132723 [00:01<00:00, 50779.77KB/s]
- 84%|########4 | 111954/132723 [00:01<00:00, 56540.28KB/s]
- 89%|########9 | 118267/132723 [00:01<00:00, 52465.75KB/s]
- 95%|#########4| 125839/132723 [00:02<00:00, 58091.76KB/s]
-100%|##########| 132723/132723 [00:02<00:00, 62888.07KB/s]
+ 4%|4 | 5789/132723 [00:00<00:02, 57884.26KB/s]
+ 11%|# | 14260/132723 [00:00<00:01, 73657.72KB/s]
+ 17%|#7 | 22751/132723 [00:00<00:01, 78793.31KB/s]
+ 24%|##3 | 31310/132723 [00:00<00:01, 81472.92KB/s]
+ 30%|##9 | 39780/132723 [00:00<00:01, 82632.96KB/s]
+ 36%|###6 | 48325/132723 [00:00<00:01, 83585.43KB/s]
+ 43%|####2 | 56684/132723 [00:00<00:00, 83499.49KB/s]
+ 49%|####9 | 65134/132723 [00:00<00:00, 83814.16KB/s]
+ 55%|#####5 | 73516/132723 [00:00<00:00, 74425.53KB/s]
+ 62%|######1 | 81905/132723 [00:01<00:00, 74779.40KB/s]
+ 68%|######8 | 90329/132723 [00:01<00:00, 77444.72KB/s]
+ 74%|#######4 | 98288/132723 [00:01<00:00, 66600.20KB/s]
+ 79%|#######9 | 105269/132723 [00:01<00:00, 65115.59KB/s]
+ 86%|########5 | 113722/132723 [00:01<00:00, 70229.24KB/s]
+ 91%|#########1| 120969/132723 [00:01<00:00, 68298.71KB/s]
+ 97%|#########7| 129306/132723 [00:01<00:00, 72419.53KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 74499.20KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -524,7 +523,7 @@ 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> ( 4 minutes 8.954 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 17.961 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 69af493ec8..ce46ed31bf 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>17:21.013</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>17:58.983</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -354,43 +354,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>04:08.954</p></td>
+<td><p>04:17.961</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:49.647</p></td>
+<td><p>04:04.015</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></td>
-<td><p>02:31.029</p></td>
+<td><p>02:37.759</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>01:53.156</p></td>
+<td><p>01:49.957</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>01:20.763</p></td>
+<td><p>01:23.693</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>00:58.545</p></td>
+<td><p>00:59.738</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_adreno_tvmc.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-tvmc-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™ with tvmc Interface</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno_tvmc.py</span></code>)</p></td>
-<td><p>00:54.934</p></td>
+<td><p>00:57.007</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></td>
-<td><p>00:45.199</p></td>
+<td><p>00:47.956</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:29.648</p></td>
+<td><p>00:30.831</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:29.133</p></td>
+<td><p>00:30.060</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index 983ab586e3..660904f6a3 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -624,7 +624,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
<span class="n">module</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">params</span></a> <span class="o">=</span> <span class="n">get_mobilenet</span><span class="p">()</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip24d80f42-0c33-45cf-8eaf-9515a068bd6e 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.zip37ec2eff-08b4-4e2b-b5c9-0b8bfaa790a7 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 379bc43d43..809c080a9c 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:57.180</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:59.464</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,19 +354,19 @@
</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>00:53.201</p></td>
+<td><p>00:55.300</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>00:02.856</p></td>
+<td><p>00:02.983</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>00:01.116</p></td>
+<td><p>00:01.173</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>
-<td><p>00:00.007</p></td>
+<td><p>00:00.008</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index a0e1b42a51..f494cd2612 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -531,10 +531,10 @@ profile the execution time of each passes.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 23154us [23154us] (48.55%; 48.55%)
-FoldScaleAxis: 24538us [9us] (51.45%; 51.45%)
- FoldConstant: 24529us [1769us] (51.43%; 99.96%)
- InferType: 22760us [22760us] (47.72%; 92.79%)
+InferType: 23568us [23568us] (48.55%; 48.55%)
+FoldScaleAxis: 24980us [9us] (51.45%; 51.45%)
+ FoldConstant: 24971us [1838us] (51.44%; 99.96%)
+ InferType: 23133us [23133us] (47.65%; 92.64%)
</pre></div>
</div>
</div>
@@ -556,10 +556,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 22816us [22816us] (48.50%; 48.50%)
-FoldScaleAxis: 24224us [8us] (51.50%; 51.50%)
- FoldConstant: 24216us [1799us] (51.48%; 99.97%)
- InferType: 22417us [22417us] (47.66%; 92.57%)
+InferType: 23479us [23479us] (48.54%; 48.54%)
+FoldScaleAxis: 24896us [8us] (51.46%; 51.46%)
+ FoldConstant: 24888us [1833us] (51.45%; 99.97%)
+ InferType: 23055us [23055us] (47.66%; 92.64%)
</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 df4be734cf..3e44ebc324 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -580,7 +580,7 @@ latency of convolution.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Convolution: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 53.512065 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 53.535198 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 64159e43cb..5e743a6678 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -862,7 +862,7 @@ be able to run on our build server</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 12.237981 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 12.243171 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 1b457e7dea..c9192dfd55 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -477,8 +477,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
<span class="nb">print</span><span class="p">(</span><span class="s2">"Baseline: </span><span class="si">%f</span><span class="s2">"</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: 0.018904
-Baseline: 3.395954
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.020891
+Baseline: 3.647581
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -537,7 +537,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt1: </span><span class="si">%f</span><span class="s2">"</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: 0.318500
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.347496
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -594,7 +594,7 @@ vastly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt2: </span><span class="si">%f</span><span class="s2">"</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: 0.359258
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.369314
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -649,7 +649,7 @@ the access pattern for A matrix is more cache friendly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt3: </span><span class="si">%f</span><span class="s2">"</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: 0.120890
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.141369
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -726,7 +726,7 @@ flattening.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt4: </span><span class="si">%f</span><span class="s2">"</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: 0.110020
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.113683
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -804,7 +804,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt5: </span><span class="si">%f</span><span class="s2">"</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: 0.111328
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.115145
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -884,7 +884,7 @@ class Module:
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</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: 0.147865
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.150701
</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 ea5728938a..d0532ee3ae 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:36.023</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:38.176</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:33.011</p></td>
+<td><p>00:34.960</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>00:01.879</p></td>
+<td><p>00:02.015</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>00:01.133</p></td>
+<td><p>00:01.201</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 2ce900f99a..057e4a16fa 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:22.795</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>10:56.303</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -354,27 +354,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></td>
-<td><p>06:19.519</p></td>
+<td><p>06:45.477</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>01:46.122</p></td>
+<td><p>01:49.529</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>01:13.644</p></td>
+<td><p>01:16.077</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>00:34.304</p></td>
+<td><p>00:34.832</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></td>
-<td><p>00:14.874</p></td>
+<td><p>00:15.551</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>00:14.333</p></td>
+<td><p>00:14.837</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 c64c46befa..7239ebaae8 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
@@ -510,12 +510,12 @@ class Module:
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- blockIdx_x = T.launch_thread("blockIdx.x", 28)
- conv2d_nchw = T.allocate([14], "float32", "local")
- pad_temp_shared = T.allocate([72], "float32", "shared")
- kernel_shared = T.allocate([3072], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 64)
- conv2d_nchw_1 = T.Buffer((14,), data=conv2d_nchw, scope="local", align=32)
+ blockIdx_x = T.launch_thread("blockIdx.x", 112)
+ conv2d_nchw = T.allocate([7], "float32", "local")
+ pad_temp_shared = T.allocate([576], "float32", "shared")
+ kernel_shared = T.allocate([6144], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 32)
+ conv2d_nchw_1 = T.Buffer((7,), data=conv2d_nchw, scope="local", align=16)
conv2d_nchw_1[0] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
@@ -523,466 +523,536 @@ class Module:
conv2d_nchw_1[4] = T.float32(0)
conv2d_nchw_1[5] = T.float32(0)
conv2d_nchw_1[6] = T.float32(0)
- conv2d_nchw_1[7] = T.float32(0)
- conv2d_nchw_1[8] = T.float32(0)
- conv2d_nchw_1[9] = T.float32(0)
- conv2d_nchw_1[10] = T.float32(0)
- conv2d_nchw_1[11] = T.float32(0)
- conv2d_nchw_1[12] = T.float32(0)
- conv2d_nchw_1[13] = T.float32(0)
- for rc_outer_outer, ry_outer_outer in T.grid(64, 3):
- cse_var_2: T.int32 = rc_outer_outer * 72
- cse_var_1: T.int32 = ry_outer_outer * 3
- pad_temp_shared_1 = T.Buffer((72,), data=pad_temp_shared, scope="shared")
- with T.launch_thread("threadIdx.x", 64) as threadIdx_x_1:
- data_1 = T.Buffer((25088,), data=data.data)
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[rc_outer_outer * 392 + threadIdx_x_1 * 4 // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + threadIdx_x_1 * 4 % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 1] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 1) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 2] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 2) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 3] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 3) % 9 - 8], T.float32(0))
+ for rc_outer_outer, rx_outer_outer in T.grid(8, 3):
+ cse_var_1: T.int32 = rc_outer_outer * 3136
threadIdx_x_1 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((576,), data=pad_temp_shared, scope="shared")
+ data_1 = T.Buffer((25088,), data=data.data)
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + threadIdx_x_1 // 9 * 49 + threadIdx_x_1 % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 32] = T.if_then_else(1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 32) // 9 * 49 + (threadIdx_x_1 + 5) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 64] = T.if_then_else(1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 64) // 9 * 49 + (threadIdx_x_1 + 1) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 96] = T.if_then_else(1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 96) // 9 * 49 + (threadIdx_x_1 + 6) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 128] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 128) // 9 * 49 + (threadIdx_x_1 + 2) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 160] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 160) // 9 * 49 + (threadIdx_x_1 + 7) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 192] = T.if_then_else(1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 192) // 9 * 49 + (threadIdx_x_1 + 3) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 224) // 9 * 49 + (threadIdx_x_1 + 8) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 256] = T.if_then_else(1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 256) // 9 * 49 + (threadIdx_x_1 + 4) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 288] = T.if_then_else(1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + threadIdx_x_1 // 9 * 49 + threadIdx_x_1 % 9 * 7 + rx_outer_outer + blockIdx_x % 7 + 1560], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 320] = T.if_then_else(1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 320) // 9 * 49 + (threadIdx_x_1 + 5) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 352] = T.if_then_else(1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 352) // 9 * 49 + (threadIdx_x_1 + 1) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 384] = T.if_then_else(1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 384) // 9 * 49 + (threadIdx_x_1 + 6) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 416] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 416) // 9 * 49 + (threadIdx_x_1 + 2) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 448) // 9 * 49 + (threadIdx_x_1 + 7) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 480] = T.if_then_else(1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 480) // 9 * 49 + (threadIdx_x_1 + 3) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 512] = T.if_then_else(1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 512) // 9 * 49 + (threadIdx_x_1 + 8) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 32):
+ pad_temp_shared_1[threadIdx_x_1 + 544] = T.if_then_else(1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8 and 1 <= rx_outer_outer + blockIdx_x % 7 and rx_outer_outer + blockIdx_x % 7 < 8, data_1[cse_var_1 + (threadIdx_x_1 + 544) // 9 * 49 + (threadIdx_x_1 + 4) % 9 * 7 + rx_outer_outer + blockIdx_x % 7 - 8], T.float32(0))
+ threadIdx_x_2 = T.env_thread("threadIdx.x")
+ kernel_shared_1 = T.Buffer((6144,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 64) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 64) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 128) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 128) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 192] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 36864]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 256) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 256) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 320) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 320) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 384] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 73728]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 448) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 448) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 512) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 512) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 576] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 110592]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 640) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 640) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 704) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 704) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 768] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 147456]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 832) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 832) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 896) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 896) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 960] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 184320]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1024) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1024) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1088) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1088) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1152] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 221184]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1216) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1216) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1280) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1280) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1344] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 258048]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1408) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1408) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1472) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1472) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1536] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 294912]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1600) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1600) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1664) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1664) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1728] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 331776]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1792) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1792) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1856) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1856) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1920] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 368640]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1984) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1984) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2048) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2048) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2112] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 405504]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2176) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2176) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2240) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2240) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2304] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 442368]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2368) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2368) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2432) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2432) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2496] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 479232]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2560) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2560) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2624) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2624) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2688] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 516096]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2752) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2752) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2816) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2816) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2880] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 552960]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2944) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2944) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 3008) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 3008) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 47]
- for i1_inner, i3_inner in T.grid(2, 7):
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 32] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 64] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 192]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 96] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 288]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 128] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 384]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 160] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 480]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 192] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4608]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 224] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4704]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 256] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4800]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 288] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4896]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 320] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 4992]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 352] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 5088]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 384] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9216]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 416] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9312]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 448] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9408]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 480] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9504]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 512] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9600]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 544] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 9696]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 576] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 13824]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 608] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 13920]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 640] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14016]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 672] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14112]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 704] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14208]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 736] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 14304]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 768] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18432]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 800] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18528]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 832] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18624]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 864] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18720]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 896] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18816]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 928] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 18912]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 960] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23040]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 992] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23136]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1024] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23232]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1056] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23328]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1088] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23424]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1120] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 23520]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1152] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27648]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1184] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27744]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1216] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27840]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1248] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 27936]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1280] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 28032]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1312] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 28128]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32256]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1376] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32352]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1408] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32448]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1440] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32544]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1472] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32640]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1504] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 32736]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1536] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 36864]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 36960]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1600] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37056]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1632] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37152]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1664] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37248]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1696] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 37344]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1728] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41472]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1760] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41568]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1792] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41664]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1824] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41760]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1856] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41856]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1888] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 41952]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1920] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46080]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1952] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46176]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 1984] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46272]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2016] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46368]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2048] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46464]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2080] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 46560]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2112] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50688]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2144] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50784]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2176] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50880]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2208] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 50976]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2240] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 51072]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2272] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 51168]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2304] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55296]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2336] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55392]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2368] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55488]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2400] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55584]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2432] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55680]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2464] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 55776]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2496] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 59904]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2528] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60000]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2560] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60096]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2592] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60192]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2624] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60288]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2656] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 60384]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64512]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2720] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64608]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2752] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64704]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2784] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64800]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2816] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64896]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2848] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 64992]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2880] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69120]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2912] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69216]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2944] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69312]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 2976] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69408]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3008] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69504]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3040] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 69600]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3072] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 73728]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3104] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 73824]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3136] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 73920]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3168] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 74016]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3200] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 74112]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3232] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 74208]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3264] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78336]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3296] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78432]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3328] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78528]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3360] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78624]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3392] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78720]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3424] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 78816]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3456] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 82944]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3488] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83040]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3520] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83136]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3552] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83232]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3584] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83328]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3616] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 83424]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3648] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87552]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3680] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87648]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3712] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87744]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3744] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87840]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3776] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 87936]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3808] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 88032]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3840] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92160]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3872] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92256]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3904] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92352]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3936] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92448]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 3968] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92544]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4000] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 92640]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4032] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96768]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4064] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96864]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4096] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 96960]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4128] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 97056]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4160] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 97152]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4192] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 97248]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4224] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101376]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4256] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101472]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4288] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101568]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4320] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101664]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4352] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101760]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4384] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 101856]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4416] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 105984]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4448] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106080]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4480] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106176]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4512] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106272]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4544] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106368]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4576] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 106464]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4608] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110592]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4640] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110688]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4672] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110784]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4704] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110880]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4736] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 110976]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4768] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 111072]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4800] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115200]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4832] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115296]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4864] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115392]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4896] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115488]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4928] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115584]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4960] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 115680]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 4992] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 119808]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5024] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 119904]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5056] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120000]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5088] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120096]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5120] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120192]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5152] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 120288]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5184] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124416]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5216] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124512]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5248] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124608]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5280] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124704]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5312] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124800]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5344] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 124896]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5376] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129024]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5408] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129120]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5440] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129216]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5472] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129312]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5504] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129408]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5536] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 129504]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5568] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133632]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5600] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133728]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5632] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133824]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5664] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 133920]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5696] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 134016]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5728] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 134112]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5760] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138240]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5792] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138336]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5824] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138432]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5856] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138528]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5888] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138624]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5920] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 138720]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5952] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 142848]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 5984] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 142944]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6016] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143040]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6048] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143136]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6080] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143232]
+ with T.launch_thread(threadIdx_x_2, 32):
+ kernel_shared_1[threadIdx_x_2 + 6112] = kernel_1[blockIdx_x // 7 * 147456 + rc_outer_outer * 576 + threadIdx_x_2 * 3 + rx_outer_outer + 143328]
+ for rc_outer_inner, yy_outer_inner in T.grid(2, 7):
+ cse_var_2: T.int32 = rc_outer_inner * 288 + yy_outer_inner
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 1] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 1]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 2] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 2]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 9] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 3]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 10] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 4]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 11] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 5]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 18] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 6]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 19] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 7]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 20] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 8]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 27] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 9]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 28] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 10]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 29] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 11]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 36] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 12]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 37] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 13]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 38] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 14]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 45] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 15]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 46] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 16]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 47] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 17]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 54] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 18]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 55] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 19]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 56] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 20]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 63] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 21]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 64] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 22]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 65] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 23]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 72] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 24]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 73] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 25]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 74] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 26]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 81] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 27]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 82] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 28]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 83] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 29]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 90] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 30]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 91] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 31]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 92] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 32]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 99] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 33]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 100] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 34]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 101] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 35]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 108] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 36]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 109] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 37]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 110] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 38]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 117] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 39]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 118] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 40]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 119] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 41]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 126] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 42]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 127] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 43]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 128] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 44]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 135] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 45]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 136] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 46]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 137] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 47]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 144] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 48]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 145] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 49]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 146] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 50]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 153] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 51]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 154] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 52]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 155] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 53]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 162] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 54]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 163] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 55]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 164] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 56]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 171] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 57]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 172] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 58]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 173] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 59]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 180] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 60]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 181] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 61]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 182] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 62]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 189] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 63]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 190] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 64]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 191] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 65]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 198] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 66]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 199] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 67]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 200] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 68]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 207] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 69]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 208] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 70]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 209] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 71]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 216] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 72]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 217] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 73]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 218] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 74]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 225] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 75]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 226] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 76]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 227] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 77]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 234] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 78]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 235] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 79]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 236] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 80]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 243] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 81]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 244] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 82]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 245] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 83]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 252] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 84]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 253] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 85]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 254] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 86]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 261] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 87]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 262] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 88]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 263] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 89]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 270] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 90]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 271] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 91]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 272] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 92]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 279] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 93]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 280] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 94]
+ conv2d_nchw_1[yy_outer_inner] = conv2d_nchw_1[yy_outer_inner] + pad_temp_shared_1[cse_var_2 + 281] * kernel_shared_1[threadIdx_x * 192 + rc_outer_inner * 96 + 95]
+ for i2_inner in range(7):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x // 7 * 6272 + threadIdx_x * 98 + i1_inner * 49 + blockIdx_x % 7 * 7 + i3_inner] = T.max(conv2d_nchw_1[i1_inner * 7 + i3_inner] + bias_1[blockIdx_x // 7 * 128 + threadIdx_x * 2 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x // 7 * 1568 + threadIdx_x * 49 + i2_inner * 7 + blockIdx_x % 7] = T.max(conv2d_nchw_1[i2_inner] + bias_1[blockIdx_x // 7 * 32 + threadIdx_x], T.float32(0))
</pre></div>
</div>
</div>
@@ -1016,7 +1086,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.337 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.352 ms
</pre></div>
</div>
</div>
@@ -1046,34 +1116,34 @@ conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
+conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=32)
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=1)
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_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_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_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=1)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
-conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=32)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
+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=1)
-compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, 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_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=1)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
@@ -1094,12 +1164,12 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+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=32)
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=4)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+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=32)
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, "unroll_explicit", True)
@@ -1126,10 +1196,10 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+extern "C" __global__ void __launch_bounds__(32) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[7];
+ __shared__ float pad_temp_shared[576];
+ __shared__ float kernel_shared[6144];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -1137,419 +1207,324 @@ extern "C" __global__ void __launch_bounds__(64) default_function_kern
conv2d_nchw[4] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- conv2d_nchw[7] = 0.000000e+00f;
- conv2d_nchw[8] = 0.000000e+00f;
- conv2d_nchw[9] = 0.000000e+00f;
- conv2d_nchw[10] = 0.000000e+00f;
- conv2d_nchw[11] = 0.000000e+00f;
- conv2d_nchw[12] = 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) {
+ for (int rc_outer_outer = 0; rc_outer_outer < 8; ++rc_outer_outer) {
+ for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
__syncthreads();
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((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) + 64) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 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) + 128) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 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) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((((((int)threadIdx.x) + 256) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 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) + 320) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 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) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((((((int)threadIdx.x) + 448) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((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) + 512) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 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) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((((((int)threadIdx.x) + 640) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 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) + 704) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 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) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((((((int)threadIdx.x) + 832) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 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) + 896) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((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) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((((((int)threadIdx.x) + 1024) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 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) + 1088) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 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) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((((((int)threadIdx.x) + 1216) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 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) + 1280) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 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) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((((((int)threadIdx.x) + 1408) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 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) + 1472) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 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) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
- kernel_shared[(((((((int)threadIdx.x) + 1600) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 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) + 1664) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 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) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
- kernel_shared[(((((((int)threadIdx.x) + 1792) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 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) + 1856) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 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) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
- kernel_shared[(((((((int)threadIdx.x) + 1984) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 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) + 2048) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 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) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
- kernel_shared[(((((((int)threadIdx.x) + 2176) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 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) + 2240) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 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) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
- kernel_shared[(((((((int)threadIdx.x) + 2368) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 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) + 2432) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 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) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
- kernel_shared[(((((((int)threadIdx.x) + 2560) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 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) + 2624) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 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) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
- kernel_shared[(((((((int)threadIdx.x) + 2752) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 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) + 2816) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 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) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
- kernel_shared[(((((((int)threadIdx.x) + 2944) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 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) + 3008) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((int)threadIdx.x) % 9)) && ((((int)threadIdx.x) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 49)) + ((((int)threadIdx.x) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 32)] = (((((1 <= ((((int)threadIdx.x) + 5) % 9)) && (((((int)threadIdx.x) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 32) / 9) * 49)) + (((((int)threadIdx.x) + 5) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 64)] = (((((1 <= ((((int)threadIdx.x) + 1) % 9)) && (((((int)threadIdx.x) + 1) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 64) / 9) * 49)) + (((((int)threadIdx.x) + 1) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 96)] = (((((1 <= ((((int)threadIdx.x) + 6) % 9)) && (((((int)threadIdx.x) + 6) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 96) / 9) * 49)) + (((((int)threadIdx.x) + 6) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 128)] = (((((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 128) / 9) * 49)) + (((((int)threadIdx.x) + 2) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 160)] = (((((1 <= ((((int)threadIdx.x) + 7) % 9)) && (((((int)threadIdx.x) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 160) / 9) * 49)) + (((((int)threadIdx.x) + 7) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 192)] = (((((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 192) / 9) * 49)) + (((((int)threadIdx.x) + 3) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= ((((int)threadIdx.x) + 8) % 9)) && (((((int)threadIdx.x) + 8) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 224) / 9) * 49)) + (((((int)threadIdx.x) + 8) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 256)] = (((((1 <= ((((int)threadIdx.x) + 4) % 9)) && (((((int)threadIdx.x) + 4) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 256) / 9) * 49)) + (((((int)threadIdx.x) + 4) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 288)] = (((((1 <= (((int)threadIdx.x) % 9)) && ((((int)threadIdx.x) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 49)) + ((((int)threadIdx.x) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) + 1560)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 320)] = (((((1 <= ((((int)threadIdx.x) + 5) % 9)) && (((((int)threadIdx.x) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 320) / 9) * 49)) + (((((int)threadIdx.x) + 5) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 352)] = (((((1 <= ((((int)threadIdx.x) + 1) % 9)) && (((((int)threadIdx.x) + 1) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 352) / 9) * 49)) + (((((int)threadIdx.x) + 1) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 384)] = (((((1 <= ((((int)threadIdx.x) + 6) % 9)) && (((((int)threadIdx.x) + 6) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 384) / 9) * 49)) + (((((int)threadIdx.x) + 6) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 416)] = (((((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 416) / 9) * 49)) + (((((int)threadIdx.x) + 2) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 <= ((((int)threadIdx.x) + 7) % 9)) && (((((int)threadIdx.x) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 448) / 9) * 49)) + (((((int)threadIdx.x) + 7) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 480)] = (((((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 480) / 9) * 49)) + (((((int)threadIdx.x) + 3) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 512)] = (((((1 <= ((((int)threadIdx.x) + 8) % 9)) && (((((int)threadIdx.x) + 8) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 512) / 9) * 49)) + (((((int)threadIdx.x) + 8) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 544)] = (((((1 <= ((((int)threadIdx.x) + 4) % 9)) && (((((int)threadIdx.x) + 4) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)blockIdx.x) % 7)))) && ((rx_outer_outer + (((int)blockIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 544) / 9) * 49)) + (((((int)threadIdx.x) + 4) % 9) * 7)) + rx_outer_outer) + (((int)blockIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 32)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 64)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 192)];
+ kernel_shared[(((int)threadIdx.x) + 96)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 288)];
+ kernel_shared[(((int)threadIdx.x) + 128)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 384)];
+ kernel_shared[(((int)threadIdx.x) + 160)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 480)];
+ kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4608)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4704)];
+ kernel_shared[(((int)threadIdx.x) + 256)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4800)];
+ kernel_shared[(((int)threadIdx.x) + 288)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4896)];
+ kernel_shared[(((int)threadIdx.x) + 320)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 4992)];
+ kernel_shared[(((int)threadIdx.x) + 352)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 5088)];
+ kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9216)];
+ kernel_shared[(((int)threadIdx.x) + 416)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9312)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9408)];
+ kernel_shared[(((int)threadIdx.x) + 480)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9504)];
+ kernel_shared[(((int)threadIdx.x) + 512)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9600)];
+ kernel_shared[(((int)threadIdx.x) + 544)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 9696)];
+ kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 13824)];
+ kernel_shared[(((int)threadIdx.x) + 608)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 13920)];
+ kernel_shared[(((int)threadIdx.x) + 640)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14016)];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14112)];
+ kernel_shared[(((int)threadIdx.x) + 704)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14208)];
+ kernel_shared[(((int)threadIdx.x) + 736)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 14304)];
+ kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18432)];
+ kernel_shared[(((int)threadIdx.x) + 800)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18528)];
+ kernel_shared[(((int)threadIdx.x) + 832)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18624)];
+ kernel_shared[(((int)threadIdx.x) + 864)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18720)];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18816)];
+ kernel_shared[(((int)threadIdx.x) + 928)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 18912)];
+ kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23040)];
+ kernel_shared[(((int)threadIdx.x) + 992)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23136)];
+ kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23232)];
+ kernel_shared[(((int)threadIdx.x) + 1056)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23328)];
+ kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23424)];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 23520)];
+ kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27648)];
+ kernel_shared[(((int)threadIdx.x) + 1184)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27744)];
+ kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27840)];
+ kernel_shared[(((int)threadIdx.x) + 1248)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 27936)];
+ kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 28032)];
+ kernel_shared[(((int)threadIdx.x) + 1312)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 28128)];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 1376)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32352)];
+ kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32448)];
+ kernel_shared[(((int)threadIdx.x) + 1440)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32544)];
+ kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32640)];
+ kernel_shared[(((int)threadIdx.x) + 1504)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 32736)];
+ kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 36864)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 36960)];
+ kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37056)];
+ kernel_shared[(((int)threadIdx.x) + 1632)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37152)];
+ kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37248)];
+ kernel_shared[(((int)threadIdx.x) + 1696)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 37344)];
+ kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41472)];
+ kernel_shared[(((int)threadIdx.x) + 1760)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41568)];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41664)];
+ kernel_shared[(((int)threadIdx.x) + 1824)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41760)];
+ kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41856)];
+ kernel_shared[(((int)threadIdx.x) + 1888)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 41952)];
+ kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46080)];
+ kernel_shared[(((int)threadIdx.x) + 1952)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46176)];
+ kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46272)];
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46368)];
+ kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46464)];
+ kernel_shared[(((int)threadIdx.x) + 2080)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 46560)];
+ kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50688)];
+ kernel_shared[(((int)threadIdx.x) + 2144)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50784)];
+ kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50880)];
+ kernel_shared[(((int)threadIdx.x) + 2208)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 50976)];
+ kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 51072)];
+ kernel_shared[(((int)threadIdx.x) + 2272)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 51168)];
+ kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55296)];
+ kernel_shared[(((int)threadIdx.x) + 2336)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55392)];
+ kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55488)];
+ kernel_shared[(((int)threadIdx.x) + 2400)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55584)];
+ kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55680)];
+ kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 55776)];
+ kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 59904)];
+ kernel_shared[(((int)threadIdx.x) + 2528)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60000)];
+ kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60096)];
+ kernel_shared[(((int)threadIdx.x) + 2592)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60192)];
+ kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60288)];
+ kernel_shared[(((int)threadIdx.x) + 2656)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 60384)];
+ kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 2720)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64608)];
+ kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64704)];
+ kernel_shared[(((int)threadIdx.x) + 2784)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64800)];
+ kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64896)];
+ kernel_shared[(((int)threadIdx.x) + 2848)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 64992)];
+ kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69120)];
+ kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69216)];
+ kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69312)];
+ kernel_shared[(((int)threadIdx.x) + 2976)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69408)];
+ kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69504)];
+ kernel_shared[(((int)threadIdx.x) + 3040)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 69600)];
+ kernel_shared[(((int)threadIdx.x) + 3072)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 73728)];
+ kernel_shared[(((int)threadIdx.x) + 3104)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 73824)];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 73920)];
+ kernel_shared[(((int)threadIdx.x) + 3168)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 74016)];
+ kernel_shared[(((int)threadIdx.x) + 3200)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 74112)];
+ kernel_shared[(((int)threadIdx.x) + 3232)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 74208)];
+ kernel_shared[(((int)threadIdx.x) + 3264)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78336)];
+ kernel_shared[(((int)threadIdx.x) + 3296)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78432)];
+ kernel_shared[(((int)threadIdx.x) + 3328)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78528)];
+ kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78624)];
+ kernel_shared[(((int)threadIdx.x) + 3392)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78720)];
+ kernel_shared[(((int)threadIdx.x) + 3424)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 78816)];
+ kernel_shared[(((int)threadIdx.x) + 3456)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 82944)];
+ kernel_shared[(((int)threadIdx.x) + 3488)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83040)];
+ kernel_shared[(((int)threadIdx.x) + 3520)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83136)];
+ kernel_shared[(((int)threadIdx.x) + 3552)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83232)];
+ kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83328)];
+ kernel_shared[(((int)threadIdx.x) + 3616)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 83424)];
+ kernel_shared[(((int)threadIdx.x) + 3648)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87552)];
+ kernel_shared[(((int)threadIdx.x) + 3680)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87648)];
+ kernel_shared[(((int)threadIdx.x) + 3712)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87744)];
+ kernel_shared[(((int)threadIdx.x) + 3744)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87840)];
+ kernel_shared[(((int)threadIdx.x) + 3776)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 87936)];
+ kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 88032)];
+ kernel_shared[(((int)threadIdx.x) + 3840)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92160)];
+ kernel_shared[(((int)threadIdx.x) + 3872)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92256)];
+ kernel_shared[(((int)threadIdx.x) + 3904)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92352)];
+ kernel_shared[(((int)threadIdx.x) + 3936)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92448)];
+ kernel_shared[(((int)threadIdx.x) + 3968)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92544)];
+ kernel_shared[(((int)threadIdx.x) + 4000)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 92640)];
+ kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96768)];
+ kernel_shared[(((int)threadIdx.x) + 4064)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96864)];
+ kernel_shared[(((int)threadIdx.x) + 4096)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 96960)];
+ kernel_shared[(((int)threadIdx.x) + 4128)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 97056)];
+ kernel_shared[(((int)threadIdx.x) + 4160)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 97152)];
+ kernel_shared[(((int)threadIdx.x) + 4192)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 97248)];
+ kernel_shared[(((int)threadIdx.x) + 4224)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101376)];
+ kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101472)];
+ kernel_shared[(((int)threadIdx.x) + 4288)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101568)];
+ kernel_shared[(((int)threadIdx.x) + 4320)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101664)];
+ kernel_shared[(((int)threadIdx.x) + 4352)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101760)];
+ kernel_shared[(((int)threadIdx.x) + 4384)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 101856)];
+ kernel_shared[(((int)threadIdx.x) + 4416)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 105984)];
+ kernel_shared[(((int)threadIdx.x) + 4448)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106080)];
+ kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106176)];
+ kernel_shared[(((int)threadIdx.x) + 4512)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106272)];
+ kernel_shared[(((int)threadIdx.x) + 4544)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106368)];
+ kernel_shared[(((int)threadIdx.x) + 4576)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 106464)];
+ kernel_shared[(((int)threadIdx.x) + 4608)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110592)];
+ kernel_shared[(((int)threadIdx.x) + 4640)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110688)];
+ kernel_shared[(((int)threadIdx.x) + 4672)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110784)];
+ kernel_shared[(((int)threadIdx.x) + 4704)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110880)];
+ kernel_shared[(((int)threadIdx.x) + 4736)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 110976)];
+ kernel_shared[(((int)threadIdx.x) + 4768)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 111072)];
+ kernel_shared[(((int)threadIdx.x) + 4800)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115200)];
+ kernel_shared[(((int)threadIdx.x) + 4832)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115296)];
+ kernel_shared[(((int)threadIdx.x) + 4864)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115392)];
+ kernel_shared[(((int)threadIdx.x) + 4896)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115488)];
+ kernel_shared[(((int)threadIdx.x) + 4928)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115584)];
+ kernel_shared[(((int)threadIdx.x) + 4960)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 115680)];
+ kernel_shared[(((int)threadIdx.x) + 4992)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 119808)];
+ kernel_shared[(((int)threadIdx.x) + 5024)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 119904)];
+ kernel_shared[(((int)threadIdx.x) + 5056)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120000)];
+ kernel_shared[(((int)threadIdx.x) + 5088)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120096)];
+ kernel_shared[(((int)threadIdx.x) + 5120)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120192)];
+ kernel_shared[(((int)threadIdx.x) + 5152)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 120288)];
+ kernel_shared[(((int)threadIdx.x) + 5184)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124416)];
+ kernel_shared[(((int)threadIdx.x) + 5216)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124512)];
+ kernel_shared[(((int)threadIdx.x) + 5248)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124608)];
+ kernel_shared[(((int)threadIdx.x) + 5280)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124704)];
+ kernel_shared[(((int)threadIdx.x) + 5312)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124800)];
+ kernel_shared[(((int)threadIdx.x) + 5344)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 124896)];
+ kernel_shared[(((int)threadIdx.x) + 5376)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129024)];
+ kernel_shared[(((int)threadIdx.x) + 5408)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129120)];
+ kernel_shared[(((int)threadIdx.x) + 5440)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129216)];
+ kernel_shared[(((int)threadIdx.x) + 5472)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129312)];
+ kernel_shared[(((int)threadIdx.x) + 5504)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129408)];
+ kernel_shared[(((int)threadIdx.x) + 5536)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 129504)];
+ kernel_shared[(((int)threadIdx.x) + 5568)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133632)];
+ kernel_shared[(((int)threadIdx.x) + 5600)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133728)];
+ kernel_shared[(((int)threadIdx.x) + 5632)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133824)];
+ kernel_shared[(((int)threadIdx.x) + 5664)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 133920)];
+ kernel_shared[(((int)threadIdx.x) + 5696)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 134016)];
+ kernel_shared[(((int)threadIdx.x) + 5728)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 134112)];
+ kernel_shared[(((int)threadIdx.x) + 5760)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138240)];
+ kernel_shared[(((int)threadIdx.x) + 5792)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138336)];
+ kernel_shared[(((int)threadIdx.x) + 5824)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138432)];
+ kernel_shared[(((int)threadIdx.x) + 5856)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138528)];
+ kernel_shared[(((int)threadIdx.x) + 5888)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138624)];
+ kernel_shared[(((int)threadIdx.x) + 5920)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 138720)];
+ kernel_shared[(((int)threadIdx.x) + 5952)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 142848)];
+ kernel_shared[(((int)threadIdx.x) + 5984)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 142944)];
+ kernel_shared[(((int)threadIdx.x) + 6016)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143040)];
+ kernel_shared[(((int)threadIdx.x) + 6048)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143136)];
+ kernel_shared[(((int)threadIdx.x) + 6080)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143232)];
+ kernel_shared[(((int)threadIdx.x) + 6112)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 576)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 143328)];
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 2; ++rc_outer_inner) {
+ for (int yy_outer_inner = 0; yy_outer_inner < 7; ++yy_outer_inner) {
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[((rc_outer_inner * 288) + yy_outer_inner)] * kernel_shared[((((int)threadIdx.x) * 192) + (rc_outer_inner * 96))]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 1)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 1)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 2)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 2)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 9)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 3)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 10)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 4)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 11)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 5)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 18)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 6)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 19)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 7)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 20)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 8)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 27)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 9)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 28)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 10)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 29)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 11)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 36)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 12)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 37)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 13)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 38)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 14)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 45)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 15)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 46)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 16)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 47)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 17)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 54)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 18)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 55)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 19)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 56)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 20)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 63)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 21)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 64)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 22)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 65)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 23)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 72)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 24)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 73)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 25)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 74)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 26)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 81)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 27)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 82)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 28)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 83)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 29)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 90)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 30)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 91)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 31)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 92)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 32)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 99)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 33)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 100)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 34)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 101)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 35)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 108)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 36)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 109)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 37)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 110)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 38)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 117)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 39)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 118)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 40)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 119)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 41)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 126)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 42)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 127)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 43)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 128)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 44)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 135)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 45)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 136)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 46)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 137)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 47)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 144)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 48)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 145)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 49)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 146)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 50)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 153)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 51)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 154)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 52)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 155)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 53)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 162)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 54)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 163)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 55)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 164)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 56)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 171)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 57)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 172)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 58)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 173)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 59)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 180)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 60)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 181)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 61)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 182)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 62)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 189)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 63)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 190)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 64)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 191)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 65)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 198)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 66)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 199)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 67)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 200)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 68)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 207)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 69)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 208)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 70)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 209)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 71)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 216)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 72)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 217)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 73)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 218)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 74)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 225)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 75)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 226)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 76)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 227)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 77)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 234)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 78)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 235)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 79)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 236)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 80)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 243)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 81)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 244)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 82)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 245)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 83)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 252)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 84)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 253)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 85)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 254)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 86)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 261)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 87)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 262)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 88)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 263)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 89)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 270)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 90)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 271)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 91)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 272)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 92)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 279)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 93)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 280)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 94)]));
+ conv2d_nchw[yy_outer_inner] = (conv2d_nchw[yy_outer_inner] + (pad_temp_shared[(((rc_outer_inner * 288) + yy_outer_inner) + 281)] * kernel_shared[(((((int)threadIdx.x) * 192) + (rc_outer_inner * 96)) + 95)]));
+ }
+ }
}
}
- for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
- }
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) / 7) * 1568) + (((int)threadIdx.x) * 49)) + (i2_inner * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[(((((int)blockIdx.x) / 7) * 32) + ((int)threadIdx.x))]), 0.000000e+00f);
}
}
</pre></div>
@@ -1584,7 +1559,7 @@ In the example below we resume the status and do more 5 trials.</p>
Get devices for measurement successfully!
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 19.519 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 45.477 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 cc5ee73838..0f6273f127 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -921,7 +921,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1154 8.1176 8.1183 8.1103 0.0036
+ 8.1331 8.1342 8.1374 8.1277 0.0040
</pre></div>
</div>
</div>
@@ -943,7 +943,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 13.644 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 16.077 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 a84b1fde72..e53e02ccce 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -940,7 +940,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 781.4774 768.0435 809.1578 767.2309 19.5758
+ 794.8575 794.8890 796.3091 793.3742 1.1984
</pre></div>
</div>
</div>
@@ -962,7 +962,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 46.122 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 49.529 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 7720fd3e76..ad60f5ccd1 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -637,26 +637,86 @@ class Module:
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- for i0_outer_i1_outer_fused in T.parallel(64):
- compute_1 = T.allocate([1024], "float32", "global")
- compute_2 = T.Buffer((1024,), data=compute_1)
- for nb_j_inner in range(2):
- for i_inner_init, j_init in T.grid(32, 16):
- compute_2[i_inner_init * 32 + nb_j_inner * 16 + j_init] = T.float32(0)
- for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner}), 32, 16):
- cse_var_1 = T.int32()
- placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_3: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
- cse_var_2: T.int32 = i_inner * 32 + nb_j_inner * 16 + j
- placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
- placeholder_7 = T.Buffer((32768,), data=placeholder.data)
- placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 8192 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- for i0_inner in range(32):
- cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 16384 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32
- compute_3 = T.Buffer((65536,), data=compute.data)
- placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_4:cse_var_4 + 32] = T.max(compute_2[i0_inner * 32:i0_inner * 32 + 32] + placeholder_5[cse_var_4:cse_var_4 + 32], T.Broadcast(T.float32(0), 32))
+ for i0_outer in T.parallel(32):
+ compute_1 = T.allocate([32], "float32", "global")
+ for i1_outer in range(64):
+ compute_2 = T.Buffer((32,), data=compute_1)
+ for i_outer_inner in range(2):
+ cse_var_2: T.int32 = i1_outer // 2
+ cse_var_1: T.int32 = i_outer_inner * 16
+ compute_2[cse_var_1] = T.float32(0)
+ compute_2[cse_var_1 + 1] = T.float32(0)
+ compute_2[cse_var_1 + 2] = T.float32(0)
+ compute_2[cse_var_1 + 3] = T.float32(0)
+ compute_2[cse_var_1 + 4] = T.float32(0)
+ compute_2[cse_var_1 + 5] = T.float32(0)
+ compute_2[cse_var_1 + 6] = T.float32(0)
+ compute_2[cse_var_1 + 7] = T.float32(0)
+ compute_2[cse_var_1 + 8] = T.float32(0)
+ compute_2[cse_var_1 + 9] = T.float32(0)
+ compute_2[cse_var_1 + 10] = T.float32(0)
+ compute_2[cse_var_1 + 11] = T.float32(0)
+ compute_2[cse_var_1 + 12] = T.float32(0)
+ compute_2[cse_var_1 + 13] = T.float32(0)
+ compute_2[cse_var_1 + 14] = T.float32(0)
+ compute_2[cse_var_1 + 15] = T.float32(0)
+ for elem_idx in range(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
+ placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
+ placeholder_7 = T.Buffer((32768,), data=placeholder.data)
+ placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ compute_2[cse_var_1] = compute_2[cse_var_1] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_3: T.int32 = cse_var_1 + 1
+ compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_4: T.int32 = cse_var_1 + 2
+ compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_5: T.int32 = cse_var_1 + 3
+ compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_6: T.int32 = cse_var_1 + 4
+ compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_7: T.int32 = cse_var_1 + 5
+ compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_8: T.int32 = cse_var_1 + 6
+ compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_9: T.int32 = cse_var_1 + 7
+ compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_10: T.int32 = cse_var_1 + 8
+ compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_11: T.int32 = cse_var_1 + 9
+ compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_12: T.int32 = cse_var_1 + 10
+ compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_13: T.int32 = cse_var_1 + 11
+ compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_14: T.int32 = cse_var_1 + 12
+ compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_15: T.int32 = cse_var_1 + 13
+ compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_16: T.int32 = cse_var_1 + 14
+ compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ cse_var_17: T.int32 = cse_var_1 + 15
+ compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + i_outer_inner * 512 + placeholder_8[placeholder_5[cse_var_2] + elem_idx] + 256], T.float32(0))
+ for i0_inner in range(4):
+ cse_var_18: T.int32 = i0_outer * 2048 + i0_inner * 512 + i1_outer * 8
+ compute_3 = T.Buffer((65536,), data=compute.data)
+ placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
+ compute_3[cse_var_18:cse_var_18 + 8] = T.max(compute_2[i0_inner * 8:i0_inner * 8 + 8] + placeholder_5[cse_var_18:cse_var_18 + 8], T.Broadcast(T.float32(0), 8))
</pre></div>
</div>
</div>
@@ -690,7 +750,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.644 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 2.009 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 91fc268a17..d6445bcc08 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:37.283</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:50.982</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,22 +354,22 @@
</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>00:37.246</p></td>
+<td><p>00:50.944</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>00:00.022</p></td>
+<td><p>00:00.023</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>
-<td><p>00:00.005</p></td>
+<td><p>00:00.006</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-even"><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>
+<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>00:00.005</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-odd"><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>
+<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></td>
<td><p>00:00.005</p></td>
<td><p>0.0 MB</p></td>
</tr>
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 5d281adac3..454f9f431b 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -695,254 +695,163 @@ Traceback (most recent call last):
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, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1310820
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3196677
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)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "/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 399, 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:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../src/runtime/rpc/rpc_module.cc:129
+ 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1012
+ 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
+ at ../src/runtime/rpc/rpc_endpoint.cc:804
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 804
+TVMError:
+---------------------------------------------------------------
+An error occurred during the execution of TVM.
+For more information, please see: https://tvm.apache.org/docs/errors.html
+---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
+
+During handling of the above exception, another exception occurred:
Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10019580
-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
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
- File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
- File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
- File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+ 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 179, 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):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ 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: 0x00007fa543335fa2
+ 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:187
+ 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:1621
2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
at ../include/tvm/runtime/packed_func.h:1217
1: Call
at ../include/tvm/runtime/packed_func.h:1213
0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 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):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10188163
-No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ 52: 0xffffffffffffffff
+ 51: _start
+ 50: __libc_start_main
+ 49: _Py_UnixMain
+ 48: 0x0000000000650da0
+ 47: 0x0000000000650afa
+ 46: _PyFunction_FastCallDict
+ 45: _PyEval_EvalCodeWithName
+ 44: _PyEval_EvalFrameDefault
+ 43: _PyFunction_FastCallKeywords
+ 42: _PyEval_EvalCodeWithName
+ 41: _PyEval_EvalFrameDefault
+ 40: _PyMethodDef_RawFastCallKeywords
+ 39: 0x0000000000546369
+ 38: _PyEval_EvalCodeWithName
+ 37: _PyEval_EvalFrameDefault
+ 36: _PyFunction_FastCallKeywords
+ 35: _PyEval_EvalCodeWithName
+ 34: _PyEval_EvalFrameDefault
+ 33: _PyFunction_FastCallDict
+ 32: _PyEval_EvalCodeWithName
+ 31: _PyEval_EvalFrameDefault
+ 30: _PyObject_FastCallDict
+ 29: 0x00000000004c06e1
+ 28: _PyFunction_FastCallDict
+ 27: _PyEval_EvalFrameDefault
+ 26: _PyMethodDescr_FastCallKeywords
+ 25: 0x00000000005dcb58
+ 24: 0x00000000005dc83f
+ 23: 0x00000000004ba127
+ 22: _PyEval_EvalFrameDefault
+ 21: _PyFunction_FastCallKeywords
+ 20: _PyEval_EvalFrameDefault
+ 19: _PyFunction_FastCall [('tile_f', [-1, 32, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3749161
+No: 3 GFLOPS: 11.81/11.81 result: MeasureResult(costs=(0.019608032166666668,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.010102033615112, timestamp=1682098553.3948522) [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8596601
+No: 4 GFLOPS: 239.17/239.17 result: MeasureResult(costs=(0.0009679504098360656,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1777567863464355, timestamp=1682098556.048608) [('tile_f', [-1, 2, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8233975
+No: 5 GFLOPS: 0.00/239.17 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
@@ -1064,8 +973,8 @@ Traceback (most recent call last):
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, 256, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3777848
-No: 5 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 512]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1411299
+No: 6 GFLOPS: 0.00/239.17 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
@@ -1187,10 +1096,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5887318
-No: 6 GFLOPS: 25.87/25.87 result: MeasureResult(costs=(0.0089472535,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.663716316223145, timestamp=1682041470.0681114) [('tile_f', [-1, 32, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3884397
-No: 7 GFLOPS: 5.97/25.87 result: MeasureResult(costs=(0.03877114475,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3159408569335938, timestamp=1682041471.1311076) [('tile_f', [-1, 1, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2080550
-No: 8 GFLOPS: 0.00/25.87 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9545468
+No: 7 GFLOPS: 0.00/239.17 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
@@ -1312,9 +1219,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1437864
-No: 9 GFLOPS: 70.75/70.75 result: MeasureResult(costs=(0.0032721330526315788,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4330008029937744, timestamp=1682041473.2224634) [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1809759
-No: 10 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9735445
+No: 8 GFLOPS: 0.00/239.17 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
@@ -1436,8 +1342,10 @@ Traceback (most recent call last):
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, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4255353
-No: 11 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7702724
+No: 9 GFLOPS: 94.19/239.17 result: MeasureResult(costs=(0.0024577285599999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9577372074127197, timestamp=1682098563.4863958) [('tile_f', [-1, 16, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4852554
+No: 10 GFLOPS: 30.39/239.17 result: MeasureResult(costs=(0.007618267571428571,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.270705223083496, timestamp=1682098564.4030895) [('tile_f', [-1, 16, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9521896
+No: 11 GFLOPS: 0.00/239.17 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
@@ -1559,9 +1467,9 @@ Traceback (most recent call last):
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, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4249560
-No: 12 GFLOPS: 26.73/70.75 result: MeasureResult(costs=(0.00866072775,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8893969058990479, timestamp=1682041474.1282947) [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,107745
-No: 13 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8679379
+No: 12 GFLOPS: 94.16/239.17 result: MeasureResult(costs=(0.002458553024390244,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6427075862884521, timestamp=1682098565.275473) [('tile_f', [-1, 1, 64, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1940563
+No: 13 GFLOPS: 0.00/239.17 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
@@ -1683,131 +1591,161 @@ Traceback (most recent call last):
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, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7949889
-No: 14 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
- func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 2, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1893611
+No: 14 GFLOPS: 0.00/239.17 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 399, 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:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../src/runtime/rpc/rpc_module.cc:129
+ 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1012
+ 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
+ at ../src/runtime/rpc/rpc_endpoint.cc:804
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 804
+TVMError:
+---------------------------------------------------------------
+An error occurred during the execution of TVM.
+For more information, please see: https://tvm.apache.org/docs/errors.html
+---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
+
+During handling of the above exception, another exception occurred:
Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:401
- 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:387
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:282
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:101
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ 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 179, 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: 0x00007fa6122a5fa2
+ 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:187
+ 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:1621
2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
at ../include/tvm/runtime/packed_func.h:1217
1: Call
at ../include/tvm/runtime/packed_func.h:1213
0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4636507
-No: 15 GFLOPS: 0.00/70.75 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, 4, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5619662
+No: 15 GFLOPS: 0.00/239.17 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
@@ -1929,8 +1867,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 128, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9836689
-No: 16 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1771785
+No: 16 GFLOPS: 0.00/239.17 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
@@ -2052,8 +1990,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 64]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,307769
-No: 17 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9908733
+No: 17 GFLOPS: 0.00/239.17 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
@@ -2175,8 +2113,8 @@ Traceback (most recent call last):
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, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6531548
-No: 18 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5735812
+No: 18 GFLOPS: 0.00/239.17 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
@@ -2298,8 +2236,8 @@ Traceback (most recent call last):
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, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7579363
-No: 19 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9485620
+No: 19 GFLOPS: 0.00/239.17 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
@@ -2421,8 +2359,8 @@ Traceback (most recent call last):
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, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4626423
-No: 20 GFLOPS: 0.00/70.75 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6724763
+No: 20 GFLOPS: 0.00/239.17 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
@@ -2544,7 +2482,7 @@ Traceback (most recent call last):
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, 16, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,333279
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9774217
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2583,9 +2521,9 @@ 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:
-[('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1809759
+[('tile_f', [-1, 2, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8233975
Finish loading 20 records
-Time cost of this operator: 0.002993
+Time cost of this operator: 0.001260
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autotvm-tune-conv2d-cuda-py">
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index df3e5017a3..46078b8561 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -649,10 +649,10 @@ the tuned operator.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 315.6 98.667 (1, 2, 10, 10, 3) 2 1 [315.6]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.312 1.035 (1, 6, 10, 10) 1 1 [3.312]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.952 0.298 (1, 1, 10, 10, 3) 1 1 [0.952]
-Total_time - 319.864 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 317.2 98.743 (1, 2, 10, 10, 3) 2 1 [317.2]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.067 0.955 (1, 6, 10, 10) 1 1 [3.067]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.971 0.302 (1, 1, 10, 10, 3) 1 1 [0.971]
+Total_time - 321.238 - - - - -
</pre></div>
</div>
</div>
@@ -704,13 +704,13 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 101.4 97.41 (1, 6, 10, 10, 1) 2 1 [101.4]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.74 1.672 (1, 6, 10, 10) 1 1 [1.74]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.956 0.918 (1, 1, 10, 10, 3) 1 1 [0.956]
-Total_time - 104.096 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 103.5 97.516 (1, 6, 10, 10, 1) 2 1 [103.5]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.793 1.689 (1, 6, 10, 10) 1 1 [1.793]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.843 0.795 (1, 3, 10, 10, 1) 1 1 [0.843]
+Total_time - 106.136 - - - - -
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 27.079 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 32.563 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_autotune.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_pytorch.html b/docs/how_to/work_with_microtvm/micro_pytorch.html
index b134fe8559..8bf37053e7 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -460,8 +460,7 @@ download a cat image and preprocess it to use as the model input.</p>
Downloading: "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, 12.4MB/s]
-100%|##########| 3.42M/3.42M [00:00<00:00, 19.7MB/s]
+100%|##########| 3.42M/3.42M [00:00<00:00, 92.2MB/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.
@@ -587,7 +586,7 @@ via the host <cite>main.cc`</cite> or if a Zephyr emulated board is selected as
Torch top-1 id: 282, class name: tiger cat
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 21.805 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 25.799 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-pytorch-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/12b9ecc04c41abaa12022061771821d1/micro_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index f20693458b..8f51e7837b 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -528,7 +528,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
<a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">"</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>'/tmp/tmpqjisivml/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmp1db7xoih/images/random'
</pre></div>
</div>
</div>
@@ -588,8 +588,8 @@ objects to other stuff? We can display some examples from our datasets using <co
<span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">"off"</span><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [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/tmpqjisivml/images/target contains 8144 images
-/tmp/tmpqjisivml/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp1db7xoih/images/target contains 8144 images
+/tmp/tmp1db7xoih/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -701,13 +701,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 42s - loss: 0.2136 - accuracy: 0.9228 - val_loss: 0.1013 - val_accuracy: 0.9622 - 42s/epoch - 129ms/step
+328/328 - 41s - loss: 0.2278 - accuracy: 0.9230 - val_loss: 0.1670 - val_accuracy: 0.9430 - 41s/epoch - 125ms/step
Epoch 2/3
-328/328 - 35s - loss: 0.0958 - accuracy: 0.9647 - val_loss: 0.1078 - val_accuracy: 0.9622 - 35s/epoch - 107ms/step
+328/328 - 35s - loss: 0.0970 - accuracy: 0.9642 - val_loss: 0.1137 - val_accuracy: 0.9607 - 35s/epoch - 108ms/step
Epoch 3/3
-328/328 - 35s - loss: 0.0707 - accuracy: 0.9745 - val_loss: 0.0993 - val_accuracy: 0.9668 - 35s/epoch - 107ms/step
+328/328 - 35s - loss: 0.0717 - accuracy: 0.9730 - val_loss: 0.1333 - val_accuracy: 0.9664 - 35s/epoch - 108ms/step
-<keras.callbacks.History object at 0x7fb847364fd0>
+<keras.callbacks.History object at 0x7ffa057aea10>
</pre></div>
</div>
</div>
@@ -971,7 +971,7 @@ as intended.</p>
<p>From here, we could modify the model to read live images from the camera - we have another
Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
<a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 24.422 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 32.404 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 e3390372e3..a473bd56ac 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>07:40.221</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>07:59.150</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 82%" />
@@ -354,27 +354,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">5. Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>04:24.422</p></td>
+<td><p>04:32.404</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">6. Model Tuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>01:27.079</p></td>
+<td><p>01:32.563</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_pytorch.html#sphx-glr-how-to-work-with-microtvm-micro-pytorch-py"><span class="std std-ref">4. microTVM PyTorch Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_pytorch.py</span></code>)</p></td>
-<td><p>01:21.805</p></td>
+<td><p>01:25.799</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">3. microTVM Ahead-of-Time (AOT) Compilation</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:10.761</p></td>
+<td><p>00:11.252</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_custom_ide.html#sphx-glr-how-to-work-with-microtvm-micro-custom-ide-py"><span class="std std-ref">9. Bring microTVM to your own development environment</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_custom_ide.py</span></code>)</p></td>
-<td><p>00:08.372</p></td>
+<td><p>00:09.031</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">2. microTVM TFLite Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:07.782</p></td>
+<td><p>00:08.102</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">7. Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index b7421028b6..584b3c5a95 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:38.105</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:39.848</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,19 +354,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:33.375</p></td>
+<td><p>00:35.046</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>00:03.015</p></td>
+<td><p>00:03.109</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>00:01.709</p></td>
+<td><p>00:01.685</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></td>
-<td><p>00:00.006</p></td>
+<td><p>00:00.007</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index c9211cdfee..aa67432681 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -554,7 +554,7 @@ 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">"tir.exp"</span><span class="p">,</span> <span class="n">target</span><span class="o">=</span><span class="s2">"cuda"</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><function my_cuda_math_rule at 0x7fb6f0170b00>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7ff79e34f7a0>
</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 1cd579ae62..84b76d4402 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:05.591</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:09.451</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,35 +354,35 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:02.693</p></td>
+<td><p>00:06.559</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></td>
-<td><p>00:01.328</p></td>
+<td><p>00:01.277</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></td>
-<td><p>00:00.652</p></td>
+<td><p>00:00.664</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>00:00.631</p></td>
+<td><p>00:00.643</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>00:00.134</p></td>
+<td><p>00:00.142</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
-<td><p>00:00.063</p></td>
+<td><p>00:00.068</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.058</p></td>
+<td><p>00:00.063</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></td>
-<td><p>00:00.031</p></td>
+<td><p>00:00.034</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index c495d3574c..14c6a39dab 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1622,7 +1622,7 @@ 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>
@@ -1906,7 +1906,7 @@ 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 02e8e4efca..20767b8195 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/f4b53fb6e/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index e126fc6d01..ababb5575a 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/f4b53fb6e/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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"><</span><a href="../interfaces/disposable.html" class="tsd-signature-type">Disposable</a><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = []</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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 d1cee78f9d..5a597a4f41 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/f4b53fb6e/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L357">runtime.ts:357</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L357">runtime.ts:357</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L355">runtime.ts:355</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L376">runtime.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L376">runtime.ts:376</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L367">runtime.ts:367</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L367">runtime.ts:367</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index cc9a2e5e4c..79423a910d 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/f4b53fb6e/web/src/runtime.ts#L299">runtime.ts:299</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L299">runtime.ts:299</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L297">runtime.ts:297</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L295">runtime.ts:295</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L320">runtime.ts:320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L320">runtime.ts:320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L327">runtime.ts:327</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L327">runtime.ts:327</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index 7f0e50c9c9..bd4d21d3de 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/f4b53fb6e/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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"> => </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/f4b53fb6e/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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"><</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">></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/f4b53fb6e/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = []</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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 338e4197c8..097d458978 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/f4b53fb6e/web/src/runtime.ts#L50">runtime.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L50">runtime.ts:50</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</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">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L47">runtime.ts:47</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L46">runtime.ts:46</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L45">runtime.ts:45</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">webGPUContext<span class="tsd-signature-symbol">:</span> <a href="webgpucontext.html" class="tsd-signature-type">WebGPUContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L48">runtime.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L48">runtime.ts:48</a></li>
</ul>
</aside>
</section>
@@ -203,7 +203,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L77">runtime.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L77">runtime.ts:77</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L67">runtime.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L67">runtime.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -243,7 +243,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L85">runtime.ts:85</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L85">runtime.ts:85</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <a href="cachedcallstack.html" class="tsd-signature-type">CachedCallStack</a></h4>
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L96">runtime.ts:96</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L96">runtime.ts:96</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -283,7 +283,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L73">runtime.ts:73</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L73">runtime.ts:73</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 46131b528d..795f23dbfc 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -161,7 +161,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L844">runtime.ts:844</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L844">runtime.ts:844</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</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">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L834">runtime.ts:834</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L834">runtime.ts:834</a></li>
</ul>
</aside>
</section>
@@ -234,7 +234,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L833">runtime.ts:833</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L833">runtime.ts:833</a></li>
</ul>
</aside>
</section>
@@ -251,7 +251,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L973">runtime.ts:973</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L973">runtime.ts:973</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -296,7 +296,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L932">runtime.ts:932</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -318,7 +318,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L901">runtime.ts:901</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L901">runtime.ts:901</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -381,7 +381,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -412,7 +412,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -453,7 +453,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -491,7 +491,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L922">runtime.ts:922</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L922">runtime.ts:922</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -508,7 +508,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -552,7 +552,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L943">runtime.ts:943</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L943">runtime.ts:943</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -577,7 +577,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -609,7 +609,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -640,7 +640,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -672,7 +672,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -695,7 +695,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -729,7 +729,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L986">runtime.ts:986</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L986">runtime.ts:986</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -769,7 +769,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -817,7 +817,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -857,7 +857,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -900,7 +900,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -938,7 +938,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1014,7 +1014,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1046,7 +1046,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1078,7 +1078,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1110,7 +1110,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1141,7 +1141,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L957">runtime.ts:957</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L957">runtime.ts:957</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/memory.html b/docs/reference/api/typedoc/classes/memory.html
index 80b68701c0..9d0c3746a3 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/f4b53fb6e/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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/f4b53fb6e/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/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 a2f4a1551b..0115d4d5f2 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -119,7 +119,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L614">runtime.ts:614</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L614">runtime.ts:614</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L626">runtime.ts:626</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L626">runtime.ts:626</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -186,7 +186,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L653">runtime.ts:653</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L653">runtime.ts:653</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L641">runtime.ts:641</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L641">runtime.ts:641</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L687">runtime.ts:687</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L687">runtime.ts:687</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ndarray.html b/docs/reference/api/typedoc/classes/ndarray.html
index 3bf8d8a218..b8b670e129 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/f4b53fb6e/web/src/runtime.ts#L401">runtime.ts:401</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L401">runtime.ts:401</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <a href="dldevice.html" class="tsd-signature-type">DLDevice</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L394">runtime.ts:394</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L394">runtime.ts:394</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -173,7 +173,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L390">runtime.ts:390</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L390">runtime.ts:390</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
<div class="tsd-signature tsd-kind-icon">ndim<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L388">runtime.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L388">runtime.ts:388</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
<div class="tsd-signature tsd-kind-icon">shape<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L392">runtime.ts:392</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L392">runtime.ts:392</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -225,7 +225,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L480">runtime.ts:480</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L480">runtime.ts:480</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -258,7 +258,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L524">runtime.ts:524</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L524">runtime.ts:524</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -290,7 +290,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L465">runtime.ts:465</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L465">runtime.ts:465</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -307,7 +307,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L458">runtime.ts:458</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L458">runtime.ts:458</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -339,7 +339,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L584">runtime.ts:584</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L584">runtime.ts:584</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -363,7 +363,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L553">runtime.ts:553</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L553">runtime.ts:553</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/packedfunccell.html b/docs/reference/api/typedoc/classes/packedfunccell.html
index 2c6022452a..9385894e04 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -117,7 +117,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L248">runtime.ts:248</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L255">runtime.ts:255</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L255">runtime.ts:255</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -163,7 +163,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L264">runtime.ts:264</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L264">runtime.ts:264</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index dfc29019b3..8df9dd4152 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/f4b53fb6e/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</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/f4b53fb6e/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -201,7 +201,7 @@
<div class="tsd-signature tsd-kind-icon">key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
</ul>
</aside>
</section>
@@ -211,7 +211,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/f4b53fb6e/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -242,7 +242,7 @@
<div class="tsd-signature tsd-kind-icon">socket<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">WebSocket</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
</ul>
</aside>
</section>
@@ -252,7 +252,7 @@
<div class="tsd-signature tsd-kind-icon">state<span class="tsd-signature-symbol">:</span> <a href="../enums/rpcserverstate.html" class="tsd-signature-type">RPCServerState</a><span class="tsd-signature-symbol"> = RPCServerState.InitHeader</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
</ul>
</aside>
</section>
@@ -262,7 +262,7 @@
<div class="tsd-signature tsd-kind-icon">url<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/classes/runtimecontext.html b/docs/reference/api/typedoc/classes/runtimecontext.html
index abec32bace..905278b2da 100644
--- a/docs/reference/api/typedoc/classes/runtimecontext.html
+++ b/docs/reference/api/typedoc/classes/runtimecontext.html
@@ -132,7 +132,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L148">runtime.ts:148</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L148">runtime.ts:148</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Item<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L143">runtime.ts:143</a></li>
</ul>
</aside>
</section>
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Size<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L144">runtime.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L144">runtime.ts:144</a></li>
</ul>
</aside>
</section>
@@ -192,7 +192,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Make<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L145">runtime.ts:145</a></li>
</ul>
</aside>
</section>
@@ -202,7 +202,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Sys<wbr>Lib<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L146">runtime.ts:146</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L146">runtime.ts:146</a></li>
</ul>
</aside>
</section>
@@ -219,7 +219,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L189">runtime.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -263,7 +263,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L163">runtime.ts:163</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L163">runtime.ts:163</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -280,7 +280,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L208">runtime.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L208">runtime.ts:208</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
@@ -309,7 +309,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L157">runtime.ts:157</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -326,7 +326,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L167">runtime.ts:167</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L167">runtime.ts:167</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -343,7 +343,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L198">runtime.ts:198</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index b76545156c..4b15de330d 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/f4b53fb6e/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -137,7 +137,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">value<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L233">runtime.ts:233</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L233">runtime.ts:233</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmarray.html b/docs/reference/api/typedoc/classes/tvmarray.html
index f46ce63664..80a87b52a1 100644
--- a/docs/reference/api/typedoc/classes/tvmarray.html
+++ b/docs/reference/api/typedoc/classes/tvmarray.html
@@ -133,7 +133,7 @@
<aside class="tsd-sources">
<p>Overrides <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#constructor">constructor</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L784">runtime.ts:784</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L784">runtime.ts:784</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -162,7 +162,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#ctx">ctx</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#dispose">dispose</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -197,7 +197,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L804">runtime.ts:804</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L804">runtime.ts:804</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -230,7 +230,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#gethandle">getHandle</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L796">runtime.ts:796</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L796">runtime.ts:796</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -283,7 +283,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typeindex">typeIndex</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -306,7 +306,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typekey">typeKey</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/af17cfeb4/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmobject.html b/docs/reference/api/typedoc/classes/tvmobject.html
index 73b1b9f407..b6008ab984 100644
--- a/docs/reference/api/typedoc/classes/tvmobject.html
+++ b/docs/reference/api/typedoc/classes/tvmobject.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f4b53fb6e/web/src/runtime.ts#L703">runtime.ts:703</a></li>
... 1869 lines suppressed ...