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 2023/04/07 20:23:30 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@460374fed5f805a0ceb2b0a8de07c190ab90d83f)
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 1bd20952d1 deploying docs (apache/tvm@460374fed5f805a0ceb2b0a8de07c190ab90d83f)
1bd20952d1 is described below
commit 1bd20952d13b935e0c60df63bb53041bee0ed16e
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Fri Apr 7 20:23:22 2023 +0000
deploying docs (apache/tvm@460374fed5f805a0ceb2b0a8de07c190ab90d83f)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 341525 -> 364844 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 23761 -> 24068 bytes
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 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 | 8 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 14 +-
.../tune_conv2d_layer_cuda.rst.txt | 1789 +++++++++++++++++---
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 78 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 10 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 316 +++-
.../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 | 55 +-
.../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 | 46 +-
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_pytorch.html | 9 +-
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 | 36 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 51 +-
docs/how_to/deploy_models/deploy_prequantized.html | 9 +-
.../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 | 39 +-
docs/how_to/deploy_models/sg_execution_times.html | 22 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 8 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 1789 +++++++++++++++++---
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 78 +-
.../tune_with_autotvm/sg_execution_times.html | 12 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 316 +++-
docs/how_to/work_with_microtvm/micro_autotune.html | 18 +-
docs/how_to/work_with_microtvm/micro_pytorch.html | 4 +-
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/install/nnpack.html | 12 +-
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 | 269 +--
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 24 +-
docs/tutorial/tensor_expr_get_started.html | 46 +-
130 files changed, 4650 insertions(+), 1475 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 8c332b479c..2b438e4497 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 3e49b1417a..d68a0ab65a 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 d1db4e538f..5071f33a80 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 21.056 seconds)
+ **Total running time of the script:** ( 1 minutes 23.136 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 1c6d5dd74c..8d7a98a4ca 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.zipd09201e7-5a6f-4f2d-8680-3686aa87f71c from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip2fdde9ab-eec0-4dc4-b7bb-73a56e975742 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 1c1f43298b..b528d8ff85 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]
19%|#8 | 7.88M/41.5M [00:00<00:00, 82.6MB/s]
38%|###7 | 15.8M/41.5M [00:00<00:00, 43.0MB/s]
54%|#####3 | 22.3M/41.5M [00:00<00:00, 42.6MB/s]
65%|######4 | 26.9M/41.5M [00:00<00:00, 33.1MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 32.7MB/s]
85%|########5 | 35.4M/41.5M [00:01<00:00, 29.7MB/s]
92%|#########2| 38.4M/41.5M [00:01<00:00, 29.8MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 34.8MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 68.4MB/s]
35%|###4 | 14.5M/41.5M [00:00<00:00, 57.6MB/s]
48%|####8 | 20.0M/41.5M [00:00<00:00, 50.2MB/s]
60%|#####9 | 24.9M/41.5M [00:00<00:00, 44.6MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 52.3MB/s]
96%|#########6| 40.0M/41.5M [00:00<00:00, 56.4MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 55.7MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index ce4117f60e..09cf54dc40 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]
32%|###2 | 14.5M/44.7M [00:00<00:00, 151MB/s]
65%|######4 | 28.9M/44.7M [00:00<00:00, 116MB/s]
91%|######### | 40.4M/44.7M [00:00<00:00, 109MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 113MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 69.2MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 71.3MB/s]
68%|######7 | 30.3M/44.7M [00:00<00:00, 96.6MB/s]
88%|########8 | 39.5M/44.7M [00:00<00:00, 79.8MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 67.3MB/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 1e3491fc4a..2fe4472069 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 33.607 seconds)
+ **Total running time of the script:** ( 1 minutes 33.505 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 abafa83beb..997a1c4e89 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**06:51.258** total execution time for **how_to_compile_models** files:
+**06:58.716** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:33.607 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:33.505 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:21.056 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:23.136 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:55.542 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:57.393 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:38.350 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:38.905 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:33.216 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:33.107 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:30.013 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:30.891 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:28.594 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:28.334 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:25.198 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:26.543 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:22.902 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:24.133 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.781 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.768 | 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 8ffd2a7318..872bd3afc1 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
@@ -637,7 +637,7 @@ 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)
- 2535.4974 2535.2080 2538.1975 2533.6781 1.3019
+ 2752.7754 2752.7170 2756.0316 2750.1003 1.7882
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 e637c7fa41..84c17d71e5 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
3776512/102967424 [>.............................] - ETA: 1s
10575872/102967424 [==>...........................] - ETA: 0s
16769024/102967424 [===>..........................] - ETA: 0s
25157632/102967424 [======>.......................] - ETA: 0s
33546240/102967424 [========>.....................] - ETA: 0s
41934848/102967424 [===========>..................] - ETA: 0s
50503680/102967424 [=============>................] - ETA: 0s
59097088/102967424 [================>.............] - ETA: 0s
64618496/102967424 [=================>............] - ETA: 0s
68820992/102967424 [===================>..........] - ETA: 0s
74702848/102967424 [====================>.........] - ETA: 0s
79716352/102967424 [======================>.......] - ETA: 0s
84606976/102967424 [=======================>......] - ETA: 0s
87810048/102967424 [========================>.....] - ETA: 0s
92266496/102967424 [=========================>....] -
ETA: 0s
99999744/102967424 [============================>.] - ETA: 0s
102967424/102967424 [==============================] - 1s 0us/step
+
8192/102967424 [..............................] - ETA: 0s
6635520/102967424 [>.............................] - ETA: 1s
11354112/102967424 [==>...........................] - ETA: 1s
16769024/102967424 [===>..........................] - ETA: 1s
18964480/102967424 [====>.........................] - ETA: 1s
22216704/102967424 [=====>........................] - ETA: 1s
25157632/102967424 [======>.......................] - ETA: 1s
33259520/102967424 [========>.....................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
38854656/102967424 [==========>...................] - ETA: 1s
41934848/102967424 [===========>..................] - ETA: 1s
50323456/102967424 [=============>................] - ETA: 1s
58712064/102967424 [================>.............] - ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] - ETA: 0s
76079104/102967424 [=====================>........] -
ETA: 0s
83877888/102967424 [=======================>......] - ETA: 0s
90718208/102967424 [=========================>....] - 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 4427b452c9..b35f0b63f8 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)
- 16.1580 16.0368 16.8793 15.9033 0.2722
+ 16.6141 16.7317 16.8329 16.1908 0.2325
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 2f726af01c..b5d1923296 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<00:02, 58.3MB/s]
9%|9 | 16.0M/170M [00:00<00:02, 76.1MB/s]
17%|#6 | 28.2M/170M [00:00<00:01, 87.6MB/s]
21%|##1 | 36.5M/170M [00:00<00:01, 85.9MB/s]
27%|##6 | 45.1M/170M [00:00<00:01, 87.5MB/s]
31%|###1 | 53.5M/170M [00:00<00:01, 78.2MB/s]
37%|###7 | 63.7M/170M [00:00<00:01, 86.6MB/s]
42%|####2 | 72.1M/170M [00:00<00:01, 78.0MB/s]
47%|####7 | 80.0M/170M [00:01<00:01, 71.9MB/s]
51%|#####1 | 87.1M/170M [00:01<00:01, 71.8MB/s]
55%|#####5 | 94.0M/170M [00:01<00:01, 66.7MB/s]
60%|###### | 102M/170M [00:01<00:01, 68.6MB/s]
64%|######4 | 109M/170M [00:01<00:00, 63.9MB/s]
68%|######7 | 115M/170M [00:01<00:01, 50.6MB/s]
71%|####### | 120M/170M [00:01<00:01, 39.4MB/s]
74%|#######3 | 125M/170M [00:02<00:01, 39.5MB/s]
79%|#######9 | 134M/170M [00:02<00:00, 46.6MB/s]
82%|########1 | 139M/170M [00:02<00:00, 46.4MB/s]
88%|########8 | 150M/170M [00:02<00:00, 61.1MB/s]
93%|#########3| 158M/170M [00:02<00:00, 60.4MB/s]
97%|#########6| 164M/170M [00:02<00:00, 59.0MB/s]
100%|##########| 170M/170M [00:02<00:00, 60.9MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
4%|3 | 6.30M/170M [00:00<00:02, 61.7MB/s]
7%|7 | 12.2M/170M [00:00<00:03, 54.5MB/s]
10%|# | 17.4M/170M [00:00<00:03, 40.9MB/s]
14%|#4 | 24.0M/170M [00:00<00:03, 45.1MB/s]
19%|#8 | 32.0M/170M [00:00<00:02, 53.2MB/s]
24%|##3 | 40.0M/170M [00:00<00:02, 52.4MB/s]
28%|##8 | 48.0M/170M [00:00<00:02, 58.9MB/s]
33%|###2 | 56.0M/170M [00:01<00:01, 61.4MB/s]
38%|###7 | 64.0M/170M [00:01<00:01, 60.4MB/s]
42%|####2 | 72.0M/170M [00:01<00:01, 62.1MB/s]
46%|####6 | 78.3M/170M [00:01<00:01, 58.5MB/s]
50%|##### | 85.7M/170M [00:01<00:01, 63.3MB/s]
54%|#####4 | 91.9M/170M [00:01<00:01, 51.0MB/s]
57%|#####7 | 97.2M/170M [00:01<00:01, 43.1MB/s]
60%|###### | 102M/170M [00:02<00:01, 42.1MB/s]
63%|######2 | 107M/170M [00:02<00:01, 37.0MB/s]
66%|######5 | 112M/170M [00:02<00:01, 40.2MB/s]
71%|####### | 120M/170M [00:02<00:01, 48.4MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 48.5MB/s]
79%|#######9 | 134M/170M [00:02<00:00, 50.3MB/s]
82%|########2 | 139M/170M [00:02<00:00, 43.8MB/s]
85%|########4 | 144M/170M [00:03<00:00, 33.4MB/s]
89%|########9 | 152M/170M [00:03<00:00, 36.1MB/s]
93%|#########3| 158M/170M [00:03<00:00, 39.3MB/s]
96%|#########5| 162M/170M [00:03<00:00, 35.6MB/s]
98%|#########7| 166M/170M [00:03<00:00, 36.2MB/s]
100%|##########| 170M/170M [00:03<00:00, 46.0MB/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 32.589 seconds)
+ **Total running time of the script:** ( 3 minutes 40.667 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 cbcb00cdc5..5d2f55e4d9 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, 79.7MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 98.8MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 41.0MB/s]
90%|########9 | 12.2M/13.6M [00:00<00:00, 38.3MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 42.4MB/s]
@@ -409,7 +409,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.2261 90.1194 95.6162 89.8526 0.6157
+ 90.5014 90.3875 95.4684 90.0507 0.6352
@@ -458,7 +458,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 16.221 seconds)
+ **Total running time of the script:** ( 1 minutes 17.885 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 14d4ff59d5..a5ebbad143 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)
- 119.3129 119.4781 122.7859 117.0400 0.9863
+ 120.8390 120.7592 123.9352 120.0472 0.4423
@@ -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 35.781 seconds)
+ **Total running time of the script:** ( 2 minutes 29.508 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 bc7ee791a5..84aa6b7eab 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 31.367 seconds)
+ **Total running time of the script:** ( 1 minutes 44.272 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 c732fe4564..24457b99d2 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -170,7 +170,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
1%| | 1202/132723 [00:00<00:10, 11980.25KB/s]
3%|2 | 3618/132723 [00:00<00:06, 19104.97KB/s]
6%|6 | 8612/132723 [00:00<00:03, 33160.88KB/s]
13%|#2 | 17108/132723 [00:00<00:02, 53587.86KB/s]
20%|#9 | 25895/132723 [00:00<00:01, 65941.20KB/s]
26%|##6 | 34731/132723 [00:00<00:01, 73557.54KB/s]
33%|###2 | 43645/132723 [00:00<00:01, 78649.02KB/s]
40%|###9 | 52464/132723 [00:00<00:00, 81683.19KB/s]
46%|####6 | 61317/132723 [00:00<00:00, 83821.06KB/s]
53%|#####2 | 70217/132723 [00:01<00:00, 85417.38KB/s]
60%|#####9 | 79058/132723 [00:01<00:00, 86330.36KB/s]
66%|######6 | 87898/132723 [00:01<00:00, 86958.22KB/s]
73%|#######2 | 96818/132723 [00:01<00:00, 87635.34KB/s]
80%|#######9 | 105721/132723 [00:01<00:00, 88054.31KB/s]
86%|########6 | 114706/132723 [00:01<00:00, 88590.67KB/s]
93%|#########3|
123566/132723 [00:01<00:00, 88586.83KB/s]
100%|#########9| 132503/132723 [00:01<00:00, 88821.31KB/s]
100%|##########| 132723/132723 [00:01<00:00, 77819.97KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
1%| | 1220/132723 [00:00<00:10, 12146.18KB/s]
3%|2 | 3867/132723 [00:00<00:06, 20556.05KB/s]
7%|7 | 9608/132723 [00:00<00:03, 37364.91KB/s]
13%|#3 | 17405/132723 [00:00<00:02, 53381.44KB/s]
19%|#9 | 25323/132723 [00:00<00:01, 62679.39KB/s]
25%|##4 | 33138/132723 [00:00<00:01, 67935.15KB/s]
31%|### | 40718/132723 [00:00<00:01, 70502.75KB/s]
37%|###6 | 48701/132723 [00:00<00:01, 73467.38KB/s]
43%|####2 | 56662/132723 [00:00<00:01, 75385.78KB/s]
49%|####8 | 64535/132723 [00:01<00:00, 76415.06KB/s]
55%|#####4 | 72474/132723 [00:01<00:00, 77323.26KB/s]
61%|###### | 80412/132723 [00:01<00:00, 77945.28KB/s]
67%|######6 | 88370/132723 [00:01<00:00, 78439.05KB/s]
73%|#######2 | 96329/132723 [00:01<00:00, 78784.59KB/s]
79%|#######8 | 104266/132723 [00:01<00:00, 78960.04KB/s]
85%|########4 |
112219/132723 [00:01<00:00, 79129.96KB/s]
91%|######### | 120226/132723 [00:01<00:00, 79411.10KB/s]
97%|#########6| 128205/132723 [00:01<00:00, 79523.49KB/s]
100%|##########| 132723/132723 [00:01<00:00, 71392.30KB/s]
@@ -246,7 +246,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 53.392 seconds)
+ **Total running time of the script:** ( 4 minutes 1.809 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 bfe7234b8e..1667d14c20 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,28 +5,28 @@
Computation times
=================
-**16:15.107** total execution time for **how_to_deploy_models** files:
+**16:45.389** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 03:53.392 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 04:01.809 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:32.589 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:40.667 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:35.781 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:29.508 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:31.367 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:44.272 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:16.221 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:17.885 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:54.735 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:58.168 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:50.517 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:52.682 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:42.972 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:43.631 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:28.910 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:28.632 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:28.617 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:28.129 | 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 032cde974b..ce1f1fb63b 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.zip6506fbaf-67a8-4e5d-a4c9-8a33d6ba9b1a from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip1960c2b8-74ce-4e50-8d2e-c7a2a7cf59c6 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 c1a0c8b8e3..4d02c406b7 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
=================
-**00:54.069** total execution time for **how_to_extend_tvm** files:
+**00:57.026** 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``) | 00:50.289 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:53.068 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.714 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.838 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.059 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.113 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.007 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index fe6e388631..3430fdd0de 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: 22136us [22136us] (48.53%; 48.53%)
- FoldScaleAxis: 23475us [6us] (51.47%; 51.47%)
- FoldConstant: 23469us [1682us] (51.45%; 99.97%)
- InferType: 21786us [21786us] (47.77%; 92.83%)
+ InferType: 22858us [22858us] (48.70%; 48.70%)
+ FoldScaleAxis: 24075us [8us] (51.30%; 51.30%)
+ FoldConstant: 24067us [1807us] (51.28%; 99.97%)
+ InferType: 22260us [22260us] (47.43%; 92.49%)
@@ -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: 21436us [21436us] (48.12%; 48.12%)
- FoldScaleAxis: 23112us [5us] (51.88%; 51.88%)
- FoldConstant: 23107us [1703us] (51.87%; 99.98%)
- InferType: 21404us [21404us] (48.05%; 92.63%)
+ InferType: 22459us [22459us] (48.62%; 48.62%)
+ FoldScaleAxis: 23732us [8us] (51.38%; 51.38%)
+ FoldConstant: 23724us [1810us] (51.36%; 99.96%)
+ InferType: 21914us [21914us] (47.44%; 92.37%)
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 bc4bb5fa18..8139cac36e 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: 45.412864 ms
+ Convolution: 53.518367 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 d633a00cde..d6caece6ed 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: 11.515900 ms
+ conv2d with tensor core: 12.243971 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 67071d499a..264a375ae8 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: 0.017848
- Baseline: 3.374923
+ Numpy running time: 0.018984
+ Baseline: 3.358529
@@ -227,7 +227,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.297316
+ Opt1: 0.308085
@@ -318,7 +318,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.335475
+ Opt2: 0.347758
@@ -406,7 +406,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.118474
+ Opt3: 0.116191
@@ -523,7 +523,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109288
+ Opt4: 0.108726
@@ -635,7 +635,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111118
+ Opt5: 0.110817
@@ -748,7 +748,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.146157
+ Opt6: 0.148391
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 dca4fccc99..4e7eb7ae18 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:35.195** total execution time for **how_to_optimize_operators** files:
+**00:35.585** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.188 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.595 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.847 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.904 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.160 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.087 | 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 0842b26d9d..ac8ce34908 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,18 +5,18 @@
Computation times
=================
-**09:58.930** total execution time for **how_to_tune_with_autoscheduler** files:
+**10:15.867** 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``) | 06:05.440 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:18.394 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:42.146 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:44.131 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:11.032 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:12.362 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:32.895 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:32.469 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:13.997 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:14.549 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:13.420 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:13.962 | 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 035314701e..f52ac70148 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
@@ -244,107 +244,869 @@ cooperative fetching, unrolling and operator fusion.
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
blockIdx_x = T.launch_thread("blockIdx.x", 32)
- conv2d_nchw = T.allocate([4], "float32", "local")
- pad_temp_shared = T.allocate([162], "float32", "shared")
- kernel_shared = T.allocate([288], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 196)
- conv2d_nchw_1 = T.Buffer((4,), data=conv2d_nchw, scope="local", align=16)
+ conv2d_nchw = T.allocate([28], "float32", "local")
+ pad_temp_shared = T.allocate([2592], "float32", "shared")
+ kernel_shared = T.allocate([4608], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 28)
+ conv2d_nchw_1 = T.Buffer((196,), data=conv2d_nchw, scope="local", align=32)
conv2d_nchw_1[0] = T.float32(0)
+ conv2d_nchw_1[14] = T.float32(0)
+ conv2d_nchw_1[7] = T.float32(0)
+ conv2d_nchw_1[21] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
+ conv2d_nchw_1[15] = T.float32(0)
+ conv2d_nchw_1[8] = T.float32(0)
+ conv2d_nchw_1[22] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
+ conv2d_nchw_1[16] = T.float32(0)
+ conv2d_nchw_1[9] = T.float32(0)
+ conv2d_nchw_1[23] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
- for rc_outer_outer in range(256):
- pad_temp_shared_1 = T.Buffer((162,), data=pad_temp_shared, scope="shared")
- with T.launch_thread("threadIdx.x", 196) as threadIdx_x_1:
- if T.likely(threadIdx_x_1 < 54):
- data_1 = T.Buffer((25088,), data=data.data)
- pad_temp_shared_1[threadIdx_x_1 * 3] = T.if_then_else(3 <= threadIdx_x_1 % 27 and threadIdx_x_1 % 27 < 24 and 1 <= threadIdx_x_1 % 3, data_1[rc_outer_outer * 98 + threadIdx_x_1 // 27 * 49 + threadIdx_x_1 % 27 // 3 * 7 + threadIdx_x_1 % 3 * 3 - 8], T.float32(0))
- pad_temp_shared_1[threadIdx_x_1 * 3 + 1] = T.if_then_else(3 <= threadIdx_x_1 % 27 and threadIdx_x_1 % 27 < 24, data_1[rc_outer_outer * 98 + threadIdx_x_1 // 27 * 49 + threadIdx_x_1 % 27 // 3 * 7 + threadIdx_x_1 % 3 * 3 - 7], T.float32(0))
- pad_temp_shared_1[threadIdx_x_1 * 3 + 2] = T.if_then_else(3 <= threadIdx_x_1 % 27 and threadIdx_x_1 % 27 < 24 and threadIdx_x_1 % 3 < 2, data_1[rc_outer_outer * 98 + threadIdx_x_1 // 27 * 49 + threadIdx_x_1 % 27 // 3 * 7 + threadIdx_x_1 % 3 * 3 - 6], T.float32(0))
+ conv2d_nchw_1[17] = T.float32(0)
+ conv2d_nchw_1[10] = T.float32(0)
+ conv2d_nchw_1[24] = T.float32(0)
+ conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[18] = T.float32(0)
+ conv2d_nchw_1[11] = T.float32(0)
+ conv2d_nchw_1[25] = T.float32(0)
+ conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[19] = T.float32(0)
+ conv2d_nchw_1[12] = T.float32(0)
+ conv2d_nchw_1[26] = T.float32(0)
+ conv2d_nchw_1[6] = T.float32(0)
+ conv2d_nchw_1[20] = T.float32(0)
+ conv2d_nchw_1[13] = T.float32(0)
+ conv2d_nchw_1[27] = T.float32(0)
+ for rc_outer_outer in range(16):
+ cse_var_2: T.int32 = rc_outer_outer * 1568
+ cse_var_1: T.int32 = rc_outer_outer * 288
threadIdx_x_1 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((288,), data=kernel_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((2592,), data=pad_temp_shared, scope="shared")
+ data_1 = T.Buffer((25088,), data=data.data)
+ with T.launch_thread(threadIdx_x_1, 28):
+ pad_temp_shared_1[threadIdx_x_1 * 36] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 3] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 4] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 5] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 6] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 7] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 8] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 9] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 10] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 11] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 12] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 13] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 14] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 15] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 16] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 17] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 18] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 19] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 20] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 21] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 22] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 23] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 24] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 25] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 26] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 27] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 28] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 29] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 30] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 31] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 32] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 33] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 34] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 35] = T.float32(0)
+ with T.launch_thread(threadIdx_x_1, 28):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1008] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1009] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1010] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1011] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1012] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1013] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1014] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1015] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1016] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1017] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1018] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1019] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1020] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1021] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1022] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1023] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1024] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1025] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1026] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1027] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1028] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1029] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1030] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1031] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1032] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1033] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1034] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1035] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1036] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1037] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1038] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1039] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1040] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1041] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1042] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1043] = T.float32(0)
+ with T.launch_thread(threadIdx_x_1, 28):
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2016] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2017] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 7], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2018] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 6], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2019] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 5], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2020] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 4], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2021] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 3], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2022] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 2], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2023] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 1], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2024] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2025] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2026] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1218], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2027] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1219], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2028] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1220], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2029] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1221], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2030] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1222], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2031] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1223], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2032] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1224], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2033] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2034] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2035] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 7], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2036] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 6], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2037] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 5], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2038] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 4], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2039] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 3], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2040] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 2], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2041] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 1], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2042] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2043] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2044] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 7], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2045] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 6], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2046] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 5], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2047] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 4], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2048] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 3], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2049] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 2], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2050] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 1], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2051] = T.float32(0)
+ threadIdx_x_2 = T.env_thread("threadIdx.x")
+ kernel_shared_1 = T.Buffer((4608,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_1, 196):
- kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x * 73728 + threadIdx_x_1 // 18 * 4608 + rc_outer_outer * 18 + threadIdx_x_1 % 18]
- with T.launch_thread(threadIdx_x_1, 196):
- if T.likely(threadIdx_x_1 < 92):
- kernel_shared_1[threadIdx_x_1 + 196] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 196) // 18 * 4608 + rc_outer_outer * 18 + (threadIdx_x_1 + 16) % 18 // 3 * 3 + (threadIdx_x_1 + 1) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72 + 36]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72 + 54]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 37]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 55]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 38]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 56]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 39]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 57]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 40]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 58]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 41]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 59]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 24]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 42]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 60]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 25]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 43]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 61]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 26]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 44]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 62]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 27]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 45]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 63]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 28]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 46]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 64]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 29]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 47]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 65]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 30]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 66]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 31]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 49]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 67]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 32]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 50]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 68]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 33]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 51]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 69]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 34]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 52]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 70]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 35]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 53]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 71]
- for i1_inner in range(4):
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 28] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 56] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 56]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 84] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 112] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 112]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 140] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 168] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 196] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 196]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 224] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 224]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 252] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 252]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 280] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 280) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 308] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 308) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 20]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 336] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 336) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 364] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 364) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 392] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 392) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 420] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 420) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 448] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 448) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 160]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 476] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 476) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 504] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 504) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 216]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 532] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 532) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 244]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 560] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 560) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 272) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 588] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 588) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 12]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 616] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 616) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 40]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 644] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 644) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 68]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 672] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 672) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 96]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 700] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 700) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 124]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 728] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 728) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 152]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 756] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 756) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 180]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 784] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 784) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 208]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 812] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 812) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 236]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 840] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 840) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 264) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 868] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 868) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 4]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 896] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 896) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 32]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 924] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 924) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 60]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 952] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 952) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 88]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 980] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 980) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 116]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1008] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1008) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 144]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1036] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1036) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 172]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1064] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1064) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 200]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1092] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1092) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 228]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1120] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1120) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 256]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1148] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1148) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 284) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1176] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1176) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 24]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1204] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1204) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 52]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1232] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1232) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 80]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1260] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1260) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 108]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1288] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1288) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 136]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1316] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1316) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 164]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1344) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 192]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1372] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1372) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 220]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1400] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1400) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 248]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1428] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1428) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 276) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1456] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1456) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 16]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1484] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1484) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 44]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1512] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1512) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 72]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1540] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1540) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 100]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1568) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 128]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1596] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1596) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 156]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1624] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1624) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 184]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1652] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1652) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 212]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1680] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1680) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 240]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1708] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1708) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 268) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1736] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1736) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 8]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1764] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1764) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 36]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1792] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1792) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 64]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1820] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1820) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 92]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1848] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1848) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 120]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1876] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1876) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 148]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1904] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1904) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 176]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1932] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1932) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 204]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1960] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1960) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 232]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1988] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1988) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 260]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2016] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 32256]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2044] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2044) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2072] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2072) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 56]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2100] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2100) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2128] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2128) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 112]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2156] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2156) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2184] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2184) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2212] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2212) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 196]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2240] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2240) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 224]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2268] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2268) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 252]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2296] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2296) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2324] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2324) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 20]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2352] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2352) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2380] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2380) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2408] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2408) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2436] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2436) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2464] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2464) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 160]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2492] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2492) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2520] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2520) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 216]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2548] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2548) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 244]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2576] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2576) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 272) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2604] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2604) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 12]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2632] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2632) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 40]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2660] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2660) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 68]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2688) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 96]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2716] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2716) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 124]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2744] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2744) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 152]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2772] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2772) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 180]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2800] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2800) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 208]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2828] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2828) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 236]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2856] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2856) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 264) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2884] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2884) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 4]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2912] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2912) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 32]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2940] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2940) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 60]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2968] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2968) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 88]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2996] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2996) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 116]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3024] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3024) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 144]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3052] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3052) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 172]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3080] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3080) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 200]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3108] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3108) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 228]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3136] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3136) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 256]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3164] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3164) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 284) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3192] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3192) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 24]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3220] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3220) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 52]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3248] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3248) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 80]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3276] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3276) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 108]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3304] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3304) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 136]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3332] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3332) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 164]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3360] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3360) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 192]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3388] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3388) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 220]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3416] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3416) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 248]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3444] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3444) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 276) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3472] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3472) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 16]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3500] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3500) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 44]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3528] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3528) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 72]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3556] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3556) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 100]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3584] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3584) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 128]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3612] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3612) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 156]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3640] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3640) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 184]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3668] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3668) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 212]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3696] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3696) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 240]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3724] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3724) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 268) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3752] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3752) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 8]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3780] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3780) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 36]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3808] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3808) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 64]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3836] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3836) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 92]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3864] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3864) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 120]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3892] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3892) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 148]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3920] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3920) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 176]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3948] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3948) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 204]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3976] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3976) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 232]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4004] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4004) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 260]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4032] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 64512]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4060] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4060) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4088] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4088) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 56]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4116] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4116) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4144] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4144) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 112]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4172] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4172) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4200] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4200) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4228] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4228) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 196]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4256] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4256) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 224]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4284] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4284) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 252]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4312] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4312) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4340] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4340) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 20]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4368] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4368) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4396] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4396) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4424] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4424) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4452] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4452) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4480] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4480) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 160]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4508] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4508) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4536] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4536) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 216]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4564] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4564) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 244]
+ with T.launch_thread(threadIdx_x_2, 28):
+ if T.likely(threadIdx_x_2 < 16):
+ kernel_shared_1[threadIdx_x_2 + 4592] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4592) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 272]
+ for rc_outer_inner, ry_outer_inner in T.grid(8, 3):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ for i1_inner, i2_inner in T.grid(2, 7):
+ cse_var_3: T.int32 = i1_inner * 7 + i2_inner
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x * 784 + threadIdx_x // 49 * 196 + i1_inner * 49 + threadIdx_x % 49] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 * 4 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x // 7 * 98 + i1_inner * 49 + i2_inner * 7 + threadIdx_x % 7] = T.max(conv2d_nchw_1[cse_var_3] + bias_1[blockIdx_x * 16 + threadIdx_x // 7 * 2 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x // 7 * 98 + i1_inner * 49 + i2_inner * 7 + threadIdx_x % 7 + 392] = T.max(conv2d_nchw_1[cse_var_3 + 14] + bias_1[blockIdx_x * 16 + threadIdx_x // 7 * 2 + i1_inner + 8], T.float32(0))
@@ -394,7 +1156,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.302 ms
+ Execution time of this operator: 0.309 ms
@@ -442,33 +1204,33 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
- conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=4)
+ conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=4)
- conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
+ conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
- conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
- conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
+ conv2d_nchw_yy_o_o_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=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=1)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=8)
conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
- conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
- conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+ conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
+ conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
- 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_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+ compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
+ compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=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=1)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
@@ -491,14 +1253,14 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=28)
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=3)
+ 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=36)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=28)
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", 1024)
+ 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)
CUDA source code:
@@ -523,101 +1285,734 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[4];
- __shared__ float pad_temp_shared[162];
- __shared__ float kernel_shared[288];
+ extern "C" __global__ void __launch_bounds__(28) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[28];
+ __shared__ float pad_temp_shared[2592];
+ __shared__ float kernel_shared[4608];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[21] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[22] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[16] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[23] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 256; ++rc_outer_outer) {
+ conv2d_nchw[17] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[24] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[18] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[25] = 0.000000e+00f;
+ conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[19] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[26] = 0.000000e+00f;
+ conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[20] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[27] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
__syncthreads();
- if (((int)threadIdx.x) < 54) {
- pad_temp_shared[(((int)threadIdx.x) * 3)] = ((((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) && (1 <= (((int)threadIdx.x) % 3))) ? data[(((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + ((((int)threadIdx.x) % 3) * 3)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 3) + 1)] = (((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) ? data[(((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + ((((int)threadIdx.x) % 3) * 3)) - 7)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 3) + 2)] = ((((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) && ((((int)threadIdx.x) % 3) < 2)) ? data[(((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + ((((int)threadIdx.x) % 3) * 3)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) * 36)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 3)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 4)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 5)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 6)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 7)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 8)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 9)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 10)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 11)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 12)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 13)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 14)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 15)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 16)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 17)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 18)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 19)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 20)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 21)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 22)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 23)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 24)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 25)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 26)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 27)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 28)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 29)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 30)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 31)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 32)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 33)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 34)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 35)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1008)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1009)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1010)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1011)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1012)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1013)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1014)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1015)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1016)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1017)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1018)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1019)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1020)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1021)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1022)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1023)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1024)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1025)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1026)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1027)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1028)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1029)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1030)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1031)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1032)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1033)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1034)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1035)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1036)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1037)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1038)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1039)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1040)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1041)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1042)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1043)] = 0.000000e+00f;
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2016)] = 0.000000e+00f;
}
- kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 18)) + (((int)threadIdx.x) % 18))];
- if (((int)threadIdx.x) < 92) {
- kernel_shared[(((int)threadIdx.x) + 196)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 196) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((((int)threadIdx.x) + 16) % 18) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2017)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 7)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2018)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 6)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2019)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 5)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2020)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 4)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2021)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 3)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2022)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 2)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2023)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2024)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2025)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2026)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1218)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2027)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1219)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2028)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1220)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2029)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1221)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2030)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1222)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2031)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1223)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2032)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1224)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2033)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2034)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2035)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 7)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2036)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 6)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2037)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 5)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2038)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 4)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2039)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 3)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2040)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 2)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2041)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2042)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2043)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2044)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 7)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2045)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 6)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2046)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 5)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2047)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 4)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2048)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 3)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2049)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 2)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2050)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2051)] = 0.000000e+00f;
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 28)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 56)];
+ kernel_shared[(((int)threadIdx.x) + 84)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 112)];
+ kernel_shared[(((int)threadIdx.x) + 140)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 196)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 224)];
+ kernel_shared[(((int)threadIdx.x) + 252)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 252)];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 280) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 308)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 308) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 20)];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 336) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 364)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 364) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 392) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 420)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 420) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 448) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 160)];
+ kernel_shared[(((int)threadIdx.x) + 476)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 476) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 504) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 532)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 532) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 244)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 560) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 272) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 588) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 12)];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 616) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 40)];
+ kernel_shared[(((int)threadIdx.x) + 644)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 644) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 68)];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 672) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 700)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 700) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 124)];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 728) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 152)];
+ kernel_shared[(((int)threadIdx.x) + 756)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 756) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 180)];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 208)];
+ kernel_shared[(((int)threadIdx.x) + 812)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 812) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 236)];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 840) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 264) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 868)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 868) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 4)];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 896) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 32)];
+ kernel_shared[(((int)threadIdx.x) + 924)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 924) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 60)];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 952) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 88)];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 980) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 116)];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1008) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 1036)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1036) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 172)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1064) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 200)];
+ kernel_shared[(((int)threadIdx.x) + 1092)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1092) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 228)];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1120) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 256)];
+ kernel_shared[(((int)threadIdx.x) + 1148)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1148) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 284) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1176) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 1204)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1204) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 52)];
+ kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1232) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 80)];
+ kernel_shared[(((int)threadIdx.x) + 1260)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1260) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 108)];
+ kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1288) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 136)];
+ kernel_shared[(((int)threadIdx.x) + 1316)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1316) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 164)];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1344) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 192)];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1372) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 220)];
+ kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1400) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 248)];
+ kernel_shared[(((int)threadIdx.x) + 1428)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1428) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 276) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1456) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 16)];
+ kernel_shared[(((int)threadIdx.x) + 1484)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1484) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 44)];
+ kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1512) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 1540)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1540) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 100)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1568) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 128)];
+ kernel_shared[(((int)threadIdx.x) + 1596)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1596) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 156)];
+ kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1624) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 184)];
+ kernel_shared[(((int)threadIdx.x) + 1652)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1652) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 212)];
+ kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1680) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 240)];
+ kernel_shared[(((int)threadIdx.x) + 1708)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1708) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 268) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1736) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 8)];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1764) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 36)];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1792) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64)];
+ kernel_shared[(((int)threadIdx.x) + 1820)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1820) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 92)];
+ kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1848) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 120)];
+ kernel_shared[(((int)threadIdx.x) + 1876)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1876) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 148)];
+ kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1904) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 176)];
+ kernel_shared[(((int)threadIdx.x) + 1932)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1932) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 204)];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1960) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 232)];
+ kernel_shared[(((int)threadIdx.x) + 1988)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1988) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 260)];
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 2044)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2044) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 2072)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2072) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 56)];
+ kernel_shared[(((int)threadIdx.x) + 2100)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2100) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2128) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 112)];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2156) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 2184)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2184) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 2212)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2212) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 196)];
+ kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2240) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 224)];
+ kernel_shared[(((int)threadIdx.x) + 2268)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2268) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 252)];
+ kernel_shared[(((int)threadIdx.x) + 2296)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2296) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2324)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2324) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 20)];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2352) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 2380)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2380) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 2408)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2408) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 2436)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2436) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2464) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 160)];
+ kernel_shared[(((int)threadIdx.x) + 2492)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2492) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 2520)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2520) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2548) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 244)];
+ kernel_shared[(((int)threadIdx.x) + 2576)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2576) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 272) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2604)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2604) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 12)];
+ kernel_shared[(((int)threadIdx.x) + 2632)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2632) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 40)];
+ kernel_shared[(((int)threadIdx.x) + 2660)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2660) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 68)];
+ kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2688) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 2716)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2716) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 124)];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2744) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 152)];
+ kernel_shared[(((int)threadIdx.x) + 2772)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2772) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 180)];
+ kernel_shared[(((int)threadIdx.x) + 2800)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2800) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 208)];
+ kernel_shared[(((int)threadIdx.x) + 2828)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2828) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 236)];
+ kernel_shared[(((int)threadIdx.x) + 2856)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2856) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 264) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2884)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2884) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 4)];
+ kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2912) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 32)];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2940) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 60)];
+ kernel_shared[(((int)threadIdx.x) + 2968)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2968) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 88)];
+ kernel_shared[(((int)threadIdx.x) + 2996)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2996) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 116)];
+ kernel_shared[(((int)threadIdx.x) + 3024)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3024) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 3052)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3052) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 172)];
+ kernel_shared[(((int)threadIdx.x) + 3080)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3080) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 200)];
+ kernel_shared[(((int)threadIdx.x) + 3108)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3108) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 228)];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3136) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 256)];
+ kernel_shared[(((int)threadIdx.x) + 3164)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3164) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 284) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3192)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3192) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 3220)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3220) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 52)];
+ kernel_shared[(((int)threadIdx.x) + 3248)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3248) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 80)];
+ kernel_shared[(((int)threadIdx.x) + 3276)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3276) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 108)];
+ kernel_shared[(((int)threadIdx.x) + 3304)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3304) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 136)];
+ kernel_shared[(((int)threadIdx.x) + 3332)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3332) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 164)];
+ kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3360) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 192)];
+ kernel_shared[(((int)threadIdx.x) + 3388)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3388) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 220)];
+ kernel_shared[(((int)threadIdx.x) + 3416)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3416) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 248)];
+ kernel_shared[(((int)threadIdx.x) + 3444)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3444) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 276) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3472)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3472) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 16)];
+ kernel_shared[(((int)threadIdx.x) + 3500)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3500) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 44)];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3528) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 3556)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3556) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 100)];
+ kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3584) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 128)];
+ kernel_shared[(((int)threadIdx.x) + 3612)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3612) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 156)];
+ kernel_shared[(((int)threadIdx.x) + 3640)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3640) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 184)];
+ kernel_shared[(((int)threadIdx.x) + 3668)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3668) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 212)];
+ kernel_shared[(((int)threadIdx.x) + 3696)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3696) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 240)];
+ kernel_shared[(((int)threadIdx.x) + 3724)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3724) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 268) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3752)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3752) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 8)];
+ kernel_shared[(((int)threadIdx.x) + 3780)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3780) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 36)];
+ kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3808) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64)];
+ kernel_shared[(((int)threadIdx.x) + 3836)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3836) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 92)];
+ kernel_shared[(((int)threadIdx.x) + 3864)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3864) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 120)];
+ kernel_shared[(((int)threadIdx.x) + 3892)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3892) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 148)];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3920) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 176)];
+ kernel_shared[(((int)threadIdx.x) + 3948)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3948) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 204)];
+ kernel_shared[(((int)threadIdx.x) + 3976)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3976) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 232)];
+ kernel_shared[(((int)threadIdx.x) + 4004)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4004) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 260)];
+ kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 4060)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4060) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 4088)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4088) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 56)];
+ kernel_shared[(((int)threadIdx.x) + 4116)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4116) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 4144)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4144) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 112)];
+ kernel_shared[(((int)threadIdx.x) + 4172)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4172) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 4200)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4200) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 4228)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4228) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 196)];
+ kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4256) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 224)];
+ kernel_shared[(((int)threadIdx.x) + 4284)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4284) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 252)];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4312) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4340)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4340) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 20)];
+ kernel_shared[(((int)threadIdx.x) + 4368)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4368) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 4396)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4396) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 4424)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4424) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 4452)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4452) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4480) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 160)];
+ kernel_shared[(((int)threadIdx.x) + 4508)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4508) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 4536)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4536) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 4564)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4564) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 244)];
+ if (((int)threadIdx.x) < 16) {
+ kernel_shared[(((int)threadIdx.x) + 4592)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4592) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 272)];
}
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[((((int)threadIdx.x) / 49) * 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 54)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 55)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 56)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 57)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 58)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 59)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 60)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 61)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 62)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 63)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 64)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 65)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 66)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 67)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 68)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 51)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 69)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 52)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 70)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 53)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 71)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 8; ++rc_outer_inner) {
+ for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ }
+ }
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
- compute[((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 49) * 196)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 49) * 4)) + i1_inner)]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[((((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7)) + 392)] = max((conv2d_nchw[(((i1_inner * 7) + i2_inner) + 14)] + bias[((((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 8)]), 0.000000e+00f);
+ }
}
}
@@ -677,7 +2072,7 @@ 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 5.440 seconds)
+ **Total running time of the script:** ( 6 minutes 18.394 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 7108109c8f..8a89702543 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)
- 8.1305 8.1301 8.1391 8.1222 0.0069
+ 8.1169 8.1178 8.1304 8.1025 0.0114
@@ -675,7 +675,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 11.032 seconds)
+ **Total running time of the script:** ( 1 minutes 12.362 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 d076499604..4c2dbe099f 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)
- 751.3321 751.4278 751.6015 750.9669 0.2678
+ 762.6329 762.5277 763.3661 762.0048 0.5607
@@ -694,7 +694,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 42.146 seconds)
+ **Total running time of the script:** ( 1 minutes 44.131 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 aec842bda3..97f1b445d3 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
@@ -389,74 +389,26 @@ 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(16):
- compute_1 = T.allocate([4096], "float32", "global")
- compute_2 = T.Buffer((4096,), data=compute_1)
- for i_outer_inner, nb_j_inner in T.grid(2, 2):
- for i_inner_init in range(64):
- cse_var_1: T.int32 = i_outer_inner * 2048 + i_inner_init * 32 + nb_j_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, i_inner in T.grid(T.Let(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2], where={cse_var_2: i0_outer_i1_outer_fused * 2 + nb_j_inner}), 64):
- cse_var_2 = T.int32()
+ for i0_outer_i1_outer_fused in T.parallel(32):
+ compute_1 = T.allocate([2048], "float32", "global")
+ compute_2 = T.Buffer((2048,), data=compute_1)
+ for nb_j_inner in range(2):
+ for i_inner_init, j_init in T.grid(64, 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}), 64, 16):
+ cse_var_1 = T.int32()
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_21: T.int32 = elem_idx * 16
- cse_var_20: T.int32 = i0_outer_i1_outer_fused * 2 + nb_j_inner
- cse_var_19: T.int32 = i_outer_inner * 16384 + i_inner * 256
- cse_var_18: T.int32 = i_outer_inner * 2048 + i_inner * 32 + nb_j_inner * 16
- cse_var_17: T.int32 = cse_var_18 + 9
- cse_var_16: T.int32 = cse_var_18 + 8
- cse_var_15: T.int32 = cse_var_18 + 7
- cse_var_14: T.int32 = cse_var_18 + 6
- cse_var_13: T.int32 = cse_var_18 + 5
- cse_var_12: T.int32 = cse_var_18 + 4
- cse_var_11: T.int32 = cse_var_18 + 3
- cse_var_10: T.int32 = cse_var_18 + 2
- cse_var_9: T.int32 = cse_var_18 + 15
- cse_var_8: T.int32 = cse_var_18 + 14
- cse_var_7: T.int32 = cse_var_18 + 13
- cse_var_6: T.int32 = cse_var_18 + 12
- cse_var_5: T.int32 = cse_var_18 + 11
- cse_var_4: T.int32 = cse_var_18 + 10
- cse_var_3: T.int32 = cse_var_18 + 1
+ 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_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 1] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 2] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 3] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 4] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 5] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 6] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 7] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 8] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 9] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 10] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 11] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 12] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 13] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 14] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 15] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- for i0_inner in range(128):
- cse_var_22: T.int32 = i0_inner * 512 + i0_outer_i1_outer_fused * 32
+ compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 16384 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
+ for i0_inner in range(64):
+ cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 32768 + 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_22:cse_var_22 + 32] = T.max(compute_2[i0_inner * 32:i0_inner * 32 + 32] + placeholder_5[cse_var_22:cse_var_22 + 32], T.Broadcast(T.float32(0), 32))
+ 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))
@@ -506,7 +458,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.862 ms
+ Execution time of this operator: 1.800 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 f13c5dc624..50006a8a3b 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
=================
-**00:48.277** total execution time for **how_to_tune_with_autotvm** files:
+**00:49.144** 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``) | 00:48.243 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:49.107 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.020 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_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``) | 00:00.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``) | 00:00.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.004 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index c09df4382a..101063d81c 100644
--- a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
@@ -268,8 +268,7 @@ for this template
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 84.76/84.76 result: MeasureResult(costs=(0.0027312945238095235,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.309992551803589, timestamp=1680835770.9312594) [('tile_f', [-1, 1, 64, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3717943
- No: 2 GFLOPS: 0.00/84.76 result: Traceback (most recent call last):
+ No: 1 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
@@ -391,8 +390,8 @@ 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, 32, 2, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7517548
- No: 3 GFLOPS: 0.00/84.76 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5214228
+ 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
@@ -514,9 +513,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6568130
- No: 4 GFLOPS: 21.67/84.76 result: MeasureResult(costs=(0.010683162272727273,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1216676235198975, timestamp=1680835775.0292473) [('tile_f', [-1, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3292251
- No: 5 GFLOPS: 0.00/84.76 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7951828
+ No: 3 GFLOPS: 2.90/2.90 result: MeasureResult(costs=(0.0799303665,), error_no=MeasureErrorNo.NO_ERROR, all_cost=6.104713439941406, timestamp=1680896039.276656) [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4138875
+ No: 4 GFLOPS: 0.00/2.90 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
@@ -638,9 +637,8 @@ 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, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2186973
- No: 6 GFLOPS: 135.98/135.98 result: MeasureResult(costs=(0.0017025168611111113,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5132319927215576, timestamp=1680835778.2157695) [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5271275
- No: 7 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3780549
+ No: 5 GFLOPS: 0.00/2.90 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
@@ -762,8 +760,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6674618
- No: 8 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2476650
+ No: 6 GFLOPS: 0.00/2.90 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
@@ -885,9 +883,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9799918
- No: 9 GFLOPS: 1.04/135.98 result: MeasureResult(costs=(0.222601181,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.189843654632568, timestamp=1680835782.5912874) [('tile_f', [-1, 1, 1, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2636680
- No: 10 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 2, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7635467
+ No: 7 GFLOPS: 0.00/2.90 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
@@ -1009,8 +1006,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9989524
- No: 11 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 128]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3176350
+ No: 8 GFLOPS: 0.00/2.90 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1132,8 +1129,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1547503
- No: 12 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3180035
+ No: 9 GFLOPS: 0.00/2.90 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1255,8 +1252,8 @@ 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, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9646654
- No: 13 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5995630
+ No: 10 GFLOPS: 0.00/2.90 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
@@ -1378,8 +1375,10 @@ 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, 1, 2, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3826664
- No: 14 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6795092
+ No: 11 GFLOPS: 208.66/208.66 result: MeasureResult(costs=(0.0011094419000000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.71787428855896, timestamp=1680896046.8706958) [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7009134
+ No: 12 GFLOPS: 5.45/208.66 result: MeasureResult(costs=(0.042511791,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.697552680969238, timestamp=1680896047.9853632) [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4612311
+ No: 13 GFLOPS: 0.00/208.66 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
@@ -1501,9 +1500,131 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 32, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5253292
- No: 15 GFLOPS: 2.56/135.98 result: MeasureResult(costs=(0.0903941345,), error_no=MeasureErrorNo.NO_ERROR, all_cost=6.586085557937622, timestamp=1680835794.574986) [('tile_f', [-1, 8, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4195678
- No: 16 GFLOPS: 0.00/135.98 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, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3692672
+ No: 14 GFLOPS: 0.00/208.66 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+ func = build(s, args, target=target, runtime=runtime)
+ File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+ input_mod = lower(inputs, args, name=name, binds=binds)
+ File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+ return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+ tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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
+
+ Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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, 1, 2, 256]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4278558
+ No: 15 GFLOPS: 0.00/208.66 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
res = future.result()
File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1520,9 +1641,131 @@ for this template
raise TimeoutError()
TimeoutError
- [('tile_f', [-1, 128, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8979317
- No: 17 GFLOPS: 15.41/135.98 result: MeasureResult(costs=(0.015019139555555554,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.297766923904419, timestamp=1680835798.044399) [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1429725
- No: 18 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ [('tile_f', [-1, 128, 1, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,395402
+ No: 16 GFLOPS: 0.00/208.66 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+ func = build(s, args, target=target, runtime=runtime)
+ File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+ input_mod = lower(inputs, args, name=name, binds=binds)
+ File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+ return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+ tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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
+
+ Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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, 1, 32, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8635040
+ No: 17 GFLOPS: 0.00/208.66 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
@@ -1644,8 +1887,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3245022
- No: 19 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1018204
+ No: 18 GFLOPS: 49.45/208.66 result: MeasureResult(costs=(0.00468107908,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5602219104766846, timestamp=1680896061.1268966) [('tile_f', [-1, 8, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7199189
+ No: 19 GFLOPS: 0.00/208.66 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
@@ -1767,8 +2011,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8110476
- No: 20 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8490621
+ No: 20 GFLOPS: 0.00/208.66 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
@@ -1890,7 +2134,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559451
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,401453
@@ -1945,9 +2189,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5271275
+ [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7009134
Finish loading 20 records
- Time cost of this operator: 0.002048
+ Time cost of this operator: 0.001501
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 d828e5abd7..7fd11be963 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 312.9 98.715 (1, 2, 10, 10, 3) 2 1 [312.9]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.108 0.98 (1, 6, 10, 10) 1 1 [3.108]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.965 0.304 (1, 1, 10, 10, 3) 1 1 [0.965]
- Total_time - 316.973 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 335.4 98.824 (1, 2, 10, 10, 3) 2 1 [335.4]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.039 0.895 (1, 6, 10, 10) 1 1 [3.039]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.953 0.281 (1, 1, 10, 10, 3) 1 1 [0.953]
+ Total_time - 339.392 - - - - -
@@ -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 102.6 97.511 (1, 6, 10, 10, 1) 2 1 [102.6]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.779 1.691 (1, 6, 10, 10) 1 1 [1.779]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.839 0.798 (1, 3, 10, 10, 1) 1 1 [0.839]
- Total_time - 105.219 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.4 97.304 (1, 6, 10, 10, 1) 2 1 [100.4]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.808 1.752 (1, 6, 10, 10) 1 1 [1.808]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.974 0.944 (1, 1, 10, 10, 3) 1 1 [0.974]
+ Total_time - 103.182 - - - - -
@@ -439,7 +439,7 @@ Timing the tuned program
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 23.999 seconds)
+ **Total running time of the script:** ( 1 minutes 27.079 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 9cd97d71af..9527fe635e 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]
100%|##########| 3.42M/3.42M [00:00<00:00, 191MB/s]
+
0%| | 0.00/3.42M [00:00<?, ?B/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 45.8MB/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 16.918 seconds)
+ **Total running time of the script:** ( 1 minutes 20.594 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 7b83185ce4..d9b892bcd8 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/tmp9svv0koq/images/random'
+ '/tmp/tmprep4b99f/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: [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0]
+ :alt: [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -317,8 +317,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmp9svv0koq/images/target contains 8144 images
- /tmp/tmp9svv0koq/images/random contains 5000 images
+ /tmp/tmprep4b99f/images/target contains 8144 images
+ /tmp/tmprep4b99f/images/random contains 5000 images
@@ -493,13 +493,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 41s - loss: 0.2119 - accuracy: 0.9250 - val_loss: 0.1330 - val_accuracy: 0.9566 - 41s/epoch - 125ms/step
+ 328/328 - 41s - loss: 0.2114 - accuracy: 0.9269 - val_loss: 0.1611 - val_accuracy: 0.9422 - 41s/epoch - 126ms/step
Epoch 2/3
- 328/328 - 34s - loss: 0.1017 - accuracy: 0.9620 - val_loss: 0.1659 - val_accuracy: 0.9358 - 34s/epoch - 104ms/step
+ 328/328 - 35s - loss: 0.1053 - accuracy: 0.9621 - val_loss: 0.2311 - val_accuracy: 0.9309 - 35s/epoch - 107ms/step
Epoch 3/3
- 328/328 - 34s - loss: 0.0594 - accuracy: 0.9768 - val_loss: 0.1022 - val_accuracy: 0.9687 - 34s/epoch - 105ms/step
+ 328/328 - 35s - loss: 0.0644 - accuracy: 0.9763 - val_loss: 0.1123 - val_accuracy: 0.9619 - 35s/epoch - 107ms/step
- <keras.callbacks.History object at 0x7fb0e80c2550>
+ <keras.callbacks.History object at 0x7f2b973ad150>
@@ -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 16.347 seconds)
+ **Total running time of the script:** ( 4 minutes 25.309 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 fc5034b398..10a5aae71b 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
=================
-**07:22.699** total execution time for **how_to_work_with_microtvm** files:
+**07:39.656** total execution time for **how_to_work_with_microtvm** files:
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:16.347 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:25.309 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:23.999 | 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_pytorch.py` (``micro_pytorch.py``) | 01:16.918 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:20.594 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.271 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.618 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:07.780 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:08.383 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.385 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.674 | 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 8a44ffaee7..be651e265e 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
=================
-**00:36.659** total execution time for **how_to_work_with_relay** files:
+**00:37.759** total execution time for **how_to_work_with_relay** files:
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.015 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.981 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:02.860 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:02.971 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.778 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.800 | 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 bc22af64ad..a72bac31e6 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 0x7fb111e3ba70>
+ <function my_cuda_math_rule at 0x7f27d29a0a70>
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 fa952ae703..801199e936 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
=================
-**00:07.031** total execution time for **how_to_work_with_schedules** files:
+**00:08.666** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:04.205 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:05.849 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.317 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.287 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.621 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.628 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.614 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.616 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.127 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.133 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.062 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.065 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.055 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.056 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.030 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.031 | 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 57878fb054..cd08d28ef0 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
=================
-**00:30.711** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:32.462** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:30.704 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:32.455 | 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 595ba40800..d6efcfb536 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 33.25s!
+ resnet18_v1 inference graph built in 35.30s!
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 2308b0e26d..780e737192 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 22.41s!
+ yolov3-tiny inference graph built in 23.75s!
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 25258b02b5..95e9b00eb6 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
=================
-**01:39.389** total execution time for **topic_vta_tutorials_frontend** files:
+**01:43.306** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:50.092 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:52.322 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.298 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:50.984 | 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 58299007ba..403691d028 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
=================
-**00:03.216** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.200** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.728 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.704 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.488 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.496 | 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 0039ac3ce8..e7b8b8c85e 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
=================
-**00:00.842** total execution time for **topic_vta_tutorials** files:
+**00:00.845** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.433 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.432 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.408 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.413 | 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 8de6cc1440..ec5bd9d936 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -207,13 +207,6 @@ trials, we can load the best schedule from the log file and apply it.
-.. rst-class:: sphx-glr-script-out
-
- .. code-block:: none
-
- .T
-
-
@@ -325,7 +318,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 95.390 ms
+ Execution time of this operator: 95.638 ms
@@ -441,7 +434,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 28.268 seconds)
+ **Total running time of the script:** ( 1 minutes 35.808 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 1197d54b04..5976fb8d8d 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: 9.89/9.89 result: MeasureResult(costs=(0.027149525599999995,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6769108772277832, timestamp=1680834130.7249622) [('tile_y', [-1, 1]), ('tile_x', [-1, 512])],None,90
- No: 2 GFLOPS: 11.40/11.40 result: MeasureResult(costs=(0.0235430748,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6571638584136963, timestamp=1680834132.632239) [('tile_y', [-1, 8]), ('tile_x', [-1, 256])],None,83
- No: 3 GFLOPS: 11.77/11.77 result: MeasureResult(costs=(0.0228073032,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6115727424621582, timestamp=1680834133.262344) [('tile_y', [-1, 32]), ('tile_x', [-1, 32])],None,55
- No: 4 GFLOPS: 0.51/11.77 result: MeasureResult(costs=(0.5315375130000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.763458490371704, timestamp=1680834142.03058) [('tile_y', [-1, 64]), ('tile_x', [-1, 1])],None,6
- No: 5 GFLOPS: 11.46/11.77 result: MeasureResult(costs=(0.023427041399999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6422414779663086, timestamp=1680834142.8465457) [('tile_y', [-1, 8]), ('tile_x', [-1, 512])],None,93
- No: 6 GFLOPS: 3.94/11.77 result: MeasureResult(costs=(0.0680754666,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3717706203460693, timestamp=1680834145.4708383) [('tile_y', [-1, 64]), ('tile_x', [-1, 16])],None,46
- No: 7 GFLOPS: 14.32/14.32 result: MeasureResult(costs=(0.018742460199999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6202528476715088, timestamp=1680834147.3047724) [('tile_y', [-1, 32]), ('tile_x', [-1, 64])],None,65
- No: 8 GFLOPS: 1.37/14.32 result: MeasureResult(costs=(0.1965769178,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.3806817531585693, timestamp=1680834150.7139058) [('tile_y', [-1, 2]), ('tile_x', [-1, 1])],None,1
- No: 9 GFLOPS: 3.07/14.32 result: MeasureResult(costs=(0.08731453139999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6185553073883057, timestamp=1680834152.4457955) [('tile_y', [-1, 2]), ('tile_x', [-1, 16])],None,41
- No: 10 GFLOPS: 1.96/14.32 result: MeasureResult(costs=(0.1367203644,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4015653133392334, timestamp=1680834154.8974438) [('tile_y', [-1, 8]), ('tile_x', [-1, 1])],None,3
+ No: 1 GFLOPS: 11.43/11.43 result: MeasureResult(costs=(0.023485298199999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6510219573974609, timestamp=1680894342.9441633) [('tile_y', [-1, 128]), ('tile_x', [-1, 32])],None,57
+ No: 2 GFLOPS: 2.48/11.43 result: MeasureResult(costs=(0.1084500142,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9891068935394287, timestamp=1680894344.9383104) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+ No: 3 GFLOPS: 0.95/11.43 result: MeasureResult(costs=(0.2820340688,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.757885694503784, timestamp=1680894350.9873857) [('tile_y', [-1, 32]), ('tile_x', [-1, 2])],None,15
+ No: 4 GFLOPS: 2.12/11.43 result: MeasureResult(costs=(0.1267427568,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2686822414398193, timestamp=1680894354.5537515) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
+ No: 5 GFLOPS: 10.35/11.43 result: MeasureResult(costs=(0.0259429574,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6928188800811768, timestamp=1680894355.4408333) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
+ No: 6 GFLOPS: 12.48/12.48 result: MeasureResult(costs=(0.021502060400000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.618582010269165, timestamp=1680894356.0488431) [('tile_y', [-1, 32]), ('tile_x', [-1, 128])],None,75
+ No: 7 GFLOPS: 3.35/12.48 result: MeasureResult(costs=(0.0800758992,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.553755283355713, timestamp=1680894358.8763993) [('tile_y', [-1, 64]), ('tile_x', [-1, 8])],None,36
+ No: 8 GFLOPS: 12.64/12.64 result: MeasureResult(costs=(0.0212287364,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6901347637176514, timestamp=1680894359.484465) [('tile_y', [-1, 256]), ('tile_x', [-1, 128])],None,78
+ No: 9 GFLOPS: 0.90/12.64 result: MeasureResult(costs=(0.29937886440000006,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.013580799102783, timestamp=1680894364.6161995) [('tile_y', [-1, 128]), ('tile_x', [-1, 2])],None,17
+ No: 10 GFLOPS: 2.12/12.64 result: MeasureResult(costs=(0.1266192918,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.254546880722046, timestamp=1680894366.9056275) [('tile_y', [-1, 128]), ('tile_x', [-1, 4])],None,27
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 680bf2be89..dbc936fb73 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': 518.4311554900012, 'median': 519.1672235000055, 'std': 1.8314267201220487}
+ {'mean': 515.2317357000004, 'median': 515.0500995000016, 'std': 2.6282614626980396}
@@ -545,29 +545,30 @@ 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: 16.63/ 21.42 GFLOPS | Progress: (4/20) | 9.95 s
[Task 1/25] Current/Best: 16.85/ 21.42 GFLOPS | Progress: (8/20) | 16.06 s
[Task 1/25] Current/Best: 16.95/ 21.42 GFLOPS | Progress: (12/20) | 19.49 s
[Task 1/25] Current/Best: 13.52/ 21.42 GFLOPS | Progress: (16/20) | 23.62 s
[Task 1/25] Current/Best: 7.53/ 21.42 GFLOPS | Progress: (20/20) | 26.49 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 14.22/ 14.22 GFLOPS | Progress: (4/20) | 4.84 s
[Task 2/25] Current/Best: 5.99/ 14.22 GFLOPS | Progress: (8/20) | 7.98 s
[Task 2/25] Current/Best: 2.14/ 14.22 GFLOPS | Progress: (12/20) | 9.68 s
[Task 2/25] Current/Best: 7.43/ 14.85 GFLOPS | Progress: (16/20) | 11.35 s
[Task 2/25] Current/Best: 16.46/ 22.54 GFLOPS | Progress: (20/20) | 13.00 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 9.85/ 9.85 GFLOPS | Progress: (4/20) | 5.49 s
[Task 3/25] Current/Best: 6.89/ 16.37 GFLOPS | Progress: (8/20) | 7.89 s
[Task 3/25] Current/Best: 8.89/ 16.37 GFLOPS | Progress: (12/20) | 10.18 s
[Task 3/25] Current/Best: 3.16/ 17.00 GFLOPS | Progress: (16/20) | 13.25 s
[Task 3/25] Current/Best: 6.67/ 17.00 GFLOPS | Progress: (20/20) | 16.11 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 13.99/ 14.66 GFLOPS | Progress: (4/20) | 5.68 s
[Task 4/25] Current/Best: 10.46/ 14.66 GFLOPS | Progress: (8/20) | 11.06 s
[Task 4/25] Current/Best: 17.78/ 17.78 GFLOPS | Progress: (12/20) | 14.17 s
[Task 4/25] Current/Best: 6.46/ 17.78 GFLOPS | Progress: (16/20) | 19.22 s
[Task 4/25] Current/Best: 6.44/ 17.78 GFLOPS | Progress: (20/20) | 23.94 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 13.00/ 17.46 GFLOPS | Progress: (4/20) | 5.39 s
[Task 5/25] Current/Best: 20.48/ 20.48 GFLOPS | Progress: (8/20) | 8.27 s
[Task 5/25] Current/Best: 19.10/ 20.48 GFLOPS | Progress: (12/20) | 10.48 s
[Task 5/25] Current/Best: 5.52/ 20.48 GFLOPS | Progress: (16/20) | 13.05 s
[Task 5/25] Current/Best: 15.55/ 20.48 GFLOPS | Progress: (20/20) | 14.95 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 10.39/ 17.16 GFLOPS | Progress: (4/20) | 5.10 s
[Task 6/25] Current/Best: 9.16/ 17.16 GFLOPS | Progress: (8/20) | 9.31 s
[Task 6/25] Current/Best: 17.32/ 17.32 GFLOPS | Progress: (12/20) | 12.02 s
[Task 6/25] Current/Best: 10.68/ 20.67 GFLOPS | Progress: (16/20) | 19.14 s
[Task 6/25] Current/Best: 20.90/ 20.90 GFLOPS | Progress: (20/20) | 22.70 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 18.32/ 19.94 GFLOPS | Progress: (4/20) | 5.56 s
[Task 7/25] Current/Best: 11.75/ 19.94 GFLOPS | Progress: (8/20) | 9.80 s
[Task 7/25] Current/Best: 12.53/ 19.94 GFLOPS | Progress: (12/20) | 11.93 s
[Task 7/25] Current/Best: 16.23/ 19.94 GFLOPS | Progress: (16/20) | 14.55 s
[Task 7/25] Current/Best: 11.47/ 19.94 GFLOPS | Progress: (20/20) | 16.76 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 11.54/ 14.43 GFLOPS | Progress: (4/20) | 8.26 s
[Task 8/25] Current/Best: 8.14/ 14.43 GFLOPS | Progress: (8/20) | 12.84 s
[Task 8/25] Current/Best: 5.86/ 14.43 GFLOPS | Progress: (12/20) | 17.46 s
[Task 8/25] Current/Best: 4.17/ 16.10 GFLOPS | Progress: (16/20) | 20.10 s
[Task 8/25] Current/Best: 10.25/ 16.10 GFLOPS | Progress: (20/20) | 22.49 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 11.88/ 11.88 GFLOPS | Progress: (4/20) | 9.36 s
[Task 9/25] Current/Best: 23.43/ 23.43 GFLOPS | Progress: (8/20) | 11.86 s
[Task 9/25] Current/Best: 22.65/ 23.43 GFLOPS | Progress: (12/20) | 15.12 s
[Task 9/25] Current/Best: 10.21/ 23.43 GFLOPS | Progress: (16/20) | 16.90 s
[Task 9/25] Current/Best: 12.95/ 23.43 GFLOPS | Progress: (20/20) | 25.02 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 5.29/ 14.63 GFLOPS | Progress: (4/20) | 4.78 s
[Task 10/25] Current/Best: 18.00/ 18.00 GFLOPS | Progress: (8/20) | 6.64 s
[Task 10/25] Current/Best: 14.19/ 18.00 GFLOPS | Progress: (12/20) | 8.29 s
[Task 10/25] Current/Best: 14.44/ 18.02 GFLOPS | Progress: (16/20) | 11.24 s
[Task 10/25] Current/Best: 4.16/ 18.02 GFLOPS | Progress: (20/20) | 13.06 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 15.95/ 18.65 GFLOPS | Progress: (4/20) | 5.52 s
[Task 11/25] Current/Best: 15.50/ 18.65 GFLOPS | Progress: (8/20) | 8.17 s
[Task 11/25] Current/Best: 12.68/ 19.84 GFLOPS | Progress: (12/20) | 10.47 s
[Task 11/25] Current/Best: 18.27/ 22.15 GFLOPS | Progress: (16/20) | 12.80 s
[Task 11/25] Current/Best: 11.91/ 22.15 GFLOPS | Progress: (20/20) | 17.69 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 11.64/ 13.77 GFLOPS | Progress: (4/20) | 12.45 s
[Task 12/25] Current/Best: 9.54/ 13.77 GFLOPS | Progress: (8/20) | 16.31 s
[Task 12/25] Current/Best: 13.62/ 17.74 GFLOPS | Progress: (12/20) | 18.82 s
[Task 12/25] Current/Best: 13.94/ 17.74 GFLOPS | Progress: (16/20) | 28.11 s
[Task 12/25] Current/Best: 14.32/ 17.74 GFLOPS | Progress: (20/20) | 30.92 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 12.92/ 15.75 GFLOPS | Progress: (4/20) | 5.91 s
[Task 13/25] Current/Best: 14.77/ 15.75 GFLOPS | Progress: (8/20) | 9.51 s
[Task 13/25] Current/Best: 9.93/ 15.75 GFLOPS | Progress: (12/20) | 13.74 s
[Task 13/25] Current/Best: 11.53/ 16.64 GFLOPS | Progress: (16/20) | 17.22 s
[Task 13/25] Current/Best: 12.43/ 16.64 GFLOPS | Progress: (20/20) | 19.75 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 8.58/ 13.23 GFLOPS | Progress: (4/20) | 5.21 s
[Task 14/25] Current/Best: 14.89/ 14.89 GFLOPS | Progress: (8/20) | 9.49 s
[Task 14/25] Current/Best: 18.53/ 18.53 GFLOPS | Progress: (12/20) | 11.74 s
[Task 14/25] Current/Best: 3.38/ 18.89 GFLOPS | Progress: (16/20) | 14.64 s
[Task 14/25] Current/Best: 18.02/ 18.89 GFLOPS | Progress: (20/20) | 17.03 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 14.61/ 15.34 GFLOPS | Progress: (4/20) | 6.41 s
[Task 15/25] Current/Best: 14.66/ 15.34 GFLOPS | Progress: (8/20) | 9.06 s Done.
-
[Task 15/25] Current/Best: 18.04/ 18.04 GFLOPS | Progress: (12/20) | 15.73 s
[Task 15/25] Current/Best: 11.72/ 18.04 GFLOPS | Progress: (16/20) | 18.90 s
[Task 15/25] Current/Best: 15.96/ 18.35 GFLOPS | Progress: (20/20) | 21.62 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 3.09/ 12.74 GFLOPS | Progress: (4/20) | 6.30 s
[Task 16/25] Current/Best: 17.97/ 17.97 GFLOPS | Progress: (8/20) | 8.53 s
[Task 16/25] Current/Best: 8.17/ 17.97 GFLOPS | Progress: (12/20) | 10.69 s
[Task 16/25] Current/Best: 5.91/ 20.87 GFLOPS | Progress: (16/20) | 12.49 s
[Task 16/25] Current/Best: 16.26/ 20.87 GFLOPS | Progress: (20/20) | 14.79 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 11.55/ 18.97 GFLOPS | Progress: (4/20) | 6.42 s
[Task 17/25] Current/Best: 12.08/ 18.97 GFLOPS | Progress: (8/20) | 10.06 s
[Task 17/25] Current/Best: 16.35/ 23.29 GFLOPS | Progress: (12/20) | 12.60 s
[Task 17/25] Current/Best: 14.10/ 23.29 GFLOPS | Progress: (16/20) | 15.45 s
[Task 17/25] Current/Best: 8.19/ 23.29 GFLOPS | Progress: (20/20) | 18.02 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 15.38/ 17.96 GFLOPS | Progress: (4/20) | 4.87 s
[Task 18/25] Current/Best: 6.30/ 19.53 GFLOPS | Progress: (8/20) | 11.24 s
[Task 18/25] Current/Best: 10.43/ 19.53 GFLOPS | Progress: (12/20) | 15.32 s
[Task 18/25] Current/Best: 6.43/ 20.97 GFLOPS | Progress: (16/20) | 17.77 s
[Task 18/25] Current/Best: 14.59/ 20.97 GFLOPS | Progress: (20/20) | 20.59 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 12.26/ 20.93 GFLOPS | Progress: (4/20) | 6.35 s
[Task 19/25] Current/Best: 12.10/ 21.03 GFLOPS | Progress: (8/20) | 9.68 s
[Task 19/25] Current/Best: 20.69/ 21.03 GFLOPS | Progress: (12/20) | 14.67 s
[Task 19/25] Current/Best: 7.62/ 21.03 GFLOPS | Progress: (16/20) | 18.62 s
[Task 19/25] Current/Best: 9.04/ 21.03 GFLOPS | Progress: (20/20) | 23.03 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 18.09/ 18.09 GFLOPS | Progress: (4/20) | 5.24 s
[Task 20/25] Current/Best: 14.47/ 19.35 GFLOPS | Progress: (8/20) | 8.64 s
[Task 20/25] Current/Best: 10.71/ 19.35 GFLOPS | Progress: (12/20) | 11.28 s
[Task 20/25] Current/Best: 11.06/ 19.35 GFLOPS | Progress: (16/20) | 14.78 s
[Task 20/25] Current/Best: 5.49/ 19.35 GFLOPS | Progress: (20/20) | 18.49 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 21/25] Current/Best: 16.99/ 18.44 GFLOPS | Progress: (4/20) | 4.53 s
[Task 21/25] Current/Best: 9.90/ 18.44 GFLOPS | Progress: (8/20) | 6.86 s
[Task 21/25] Current/Best: 16.49/ 18.89 GFLOPS | Progress: (12/20) | 9.72 s
[Task 21/25] Current/Best: 16.40/ 18.89 GFLOPS | Progress: (16/20) | 11.57 s
[Task 21/25] Current/Best: 7.88/ 18.89 GFLOPS | Progress: (20/20) | 14.97 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 1.56/ 6.59 GFLOPS | Progress: (4/20) | 7.59 s
[Task 22/25] Current/Best: 10.12/ 19.07 GFLOPS | Progress: (8/20) | 9.88 s
[Task 22/25] Current/Best: 2.53/ 19.07 GFLOPS | Progress: (12/20) | 12.18 s
[Task 22/25] Current/Best: 19.72/ 19.72 GFLOPS | Progress: (16/20) | 13.88 s
[Task 22/25] Current/Best: 10.66/ 19.72 GFLOPS | Progress: (20/20) | 16.28 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 6.17/ 9.70 GFLOPS | Progress: (4/20) | 6.04 s
[Task 23/25] Current/Best: 8.78/ 23.09 GFLOPS | Progress: (8/20) | 9.92 s
[Task 23/25] Current/Best: 11.74/ 23.09 GFLOPS | Progress: (12/20) | 13.00 s
[Task 23/25] Current/Best: 11.95/ 23.09 GFLOPS | Progress: (16/20) | 16.24 s
[Task 23/25] Current/Best: 8.12/ 23.09 GFLOPS | Progress: (20/20) | 20.85 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 9.30/ 9.30 GFLOPS | Progress: (4/20) | 13.73 s
[Task 24/25] Current/Best: 9.14/ 9.30 GFLOPS | Progress: (8/20) | 21.06 s
[Task 24/25] Current/Best: 6.55/ 9.30 GFLOPS | Progress: (12/20) | 31.74 s Done.
-
[Task 24/25] Current/Best: 3.95/ 9.30 GFLOPS | Progress: (16/20) | 44.35 s
[Task 24/25] Current/Best: 10.75/ 10.98 GFLOPS | Progress: (20/20) | 54.91 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 8.09/ 8.50 GFLOPS | Progress: (4/20) | 5.36 s
[Task 25/25] Current/Best: 8.36/ 8.50 GFLOPS | Progress: (8/20) | 16.33 s
[Task 25/25] Current/Best: 7.69/ 9.30 GFLOPS | Progress: (12/20) | 27.28 s
[Task 25/25] Current/Best: 9.54/ 9.54 GFLOPS | Progress: (16/20) | 34.66 s
[Task 25/25] Current/Best: 6.98/ 9.54 GFLOPS | Progress: (20/20) | 42.85 s
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 12.96/ 14.83 GFLOPS | Progress: (4/20) | 11.38 s
[Task 1/25] Current/Best: 11.18/ 17.56 GFLOPS | Progress: (8/20) | 18.11 s
[Task 1/25] Current/Best: 14.53/ 22.25 GFLOPS | Progress: (12/20) | 20.30 s
[Task 1/25] Current/Best: 22.49/ 22.49 GFLOPS | Progress: (16/20) | 23.74 s
[Task 1/25] Current/Best: 14.10/ 22.49 GFLOPS | Progress: (20/20) | 26.47 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 8.88/ 12.40 GFLOPS | Progress: (4/20) | 4.32 s
[Task 2/25] Current/Best: 12.10/ 20.17 GFLOPS | Progress: (8/20) | 5.62 s
[Task 2/25] Current/Best: 16.94/ 20.17 GFLOPS | Progress: (12/20) | 7.09 s
[Task 2/25] Current/Best: 9.10/ 20.17 GFLOPS | Progress: (16/20) | 8.82 s
[Task 2/25] Current/Best: 12.21/ 20.17 GFLOPS | Progress: (20/20) | 11.95 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 10.30/ 18.38 GFLOPS | Progress: (4/20) | 5.28 s
[Task 3/25] Current/Best: 6.84/ 21.18 GFLOPS | Progress: (8/20) | 8.24 s
[Task 3/25] Current/Best: 7.19/ 21.18 GFLOPS | Progress: (12/20) | 10.95 s
[Task 3/25] Current/Best: 17.81/ 21.18 GFLOPS | Progress: (16/20) | 13.78 s
[Task 3/25] Current/Best: 23.11/ 23.11 GFLOPS | Progress: (20/20) | 16.55 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 13.12/ 13.12 GFLOPS | Progress: (4/20) | 9.81 s
[Task 4/25] Current/Best: 17.53/ 17.53 GFLOPS | Progress: (8/20) | 11.68 s
[Task 4/25] Current/Best: 3.79/ 17.53 GFLOPS | Progress: (12/20) | 13.86 s
[Task 4/25] Current/Best: 10.52/ 19.93 GFLOPS | Progress: (16/20) | 16.97 s
[Task 4/25] Current/Best: 2.18/ 19.93 GFLOPS | Progress: (20/20) | 23.60 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 14.67/ 14.67 GFLOPS | Progress: (4/20) | 5.18 s
[Task 5/25] Current/Best: 8.88/ 22.01 GFLOPS | Progress: (8/20) | 7.57 s
[Task 5/25] Current/Best: 17.20/ 22.01 GFLOPS | Progress: (12/20) | 9.56 s
[Task 5/25] Current/Best: 13.07/ 22.01 GFLOPS | Progress: (16/20) | 11.30 s
[Task 5/25] Current/Best: 18.32/ 22.01 GFLOPS | Progress: (20/20) | 13.49 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 16.40/ 17.58 GFLOPS | Progress: (4/20) | 6.57 s
[Task 6/25] Current/Best: 18.91/ 20.47 GFLOPS | Progress: (8/20) | 8.46 s
[Task 6/25] Current/Best: 6.10/ 20.47 GFLOPS | Progress: (12/20) | 11.80 s
[Task 6/25] Current/Best: 11.71/ 20.47 GFLOPS | Progress: (16/20) | 15.22 s
[Task 6/25] Current/Best: 12.64/ 20.47 GFLOPS | Progress: (20/20) | 17.98 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 16.70/ 20.49 GFLOPS | Progress: (4/20) | 5.39 s
[Task 7/25] Current/Best: 16.05/ 20.49 GFLOPS | Progress: (8/20) | 7.78 s
[Task 7/25] Current/Best: 12.86/ 20.49 GFLOPS | Progress: (12/20) | 9.93 s
[Task 7/25] Current/Best: 18.19/ 20.49 GFLOPS | Progress: (16/20) | 12.47 s
[Task 7/25] Current/Best: 3.09/ 20.49 GFLOPS | Progress: (20/20) | 15.72 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 11.53/ 12.36 GFLOPS | Progress: (4/20) | 5.92 s
[Task 8/25] Current/Best: 12.04/ 12.94 GFLOPS | Progress: (8/20) | 9.21 s
[Task 8/25] Current/Best: 14.12/ 18.94 GFLOPS | Progress: (12/20) | 11.44 s
[Task 8/25] Current/Best: 14.26/ 18.94 GFLOPS | Progress: (16/20) | 17.26 s
[Task 8/25] Current/Best: 10.15/ 18.94 GFLOPS | Progress: (20/20) | 25.85 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 20.71/ 20.71 GFLOPS | Progress: (4/20) | 5.04 s
[Task 9/25] Current/Best: 18.48/ 20.71 GFLOPS | Progress: (8/20) | 6.91 s
[Task 9/25] Current/Best: 13.92/ 20.71 GFLOPS | Progress: (12/20) | 9.35 s
[Task 9/25] Current/Best: 11.15/ 20.71 GFLOPS | Progress: (16/20) | 14.05 s
[Task 9/25] Current/Best: 17.88/ 20.71 GFLOPS | Progress: (20/20) | 16.08 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 13.16/ 13.16 GFLOPS | Progress: (4/20) | 4.94 s
[Task 10/25] Current/Best: 4.67/ 13.63 GFLOPS | Progress: (8/20) | 7.83 s
[Task 10/25] Current/Best: 9.95/ 19.26 GFLOPS | Progress: (12/20) | 9.87 s
[Task 10/25] Current/Best: 13.05/ 19.26 GFLOPS | Progress: (16/20) | 12.18 s
[Task 10/25] Current/Best: 6.02/ 19.26 GFLOPS | Progress: (20/20) | 13.86 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 3.08/ 23.38 GFLOPS | Progress: (4/20) | 7.21 s
[Task 11/25] Current/Best: 10.90/ 23.38 GFLOPS | Progress: (8/20) | 11.06 s
[Task 11/25] Current/Best: 14.43/ 23.38 GFLOPS | Progress: (12/20) | 13.16 s
[Task 11/25] Current/Best: 16.67/ 23.38 GFLOPS | Progress: (16/20) | 15.93 s
[Task 11/25] Current/Best: 8.74/ 23.38 GFLOPS | Progress: (20/20) | 18.43 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 9.36/ 20.25 GFLOPS | Progress: (4/20) | 5.89 s
[Task 12/25] Current/Best: 2.87/ 20.25 GFLOPS | Progress: (8/20) | 9.71 s
[Task 12/25] Current/Best: 14.24/ 20.25 GFLOPS | Progress: (12/20) | 13.33 s
[Task 12/25] Current/Best: 6.05/ 20.25 GFLOPS | Progress: (16/20) | 17.39 s
[Task 12/25] Current/Best: 9.54/ 20.54 GFLOPS | Progress: (20/20) | 23.80 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 12.85/ 18.41 GFLOPS | Progress: (4/20) | 5.50 s
[Task 13/25] Current/Best: 12.55/ 18.41 GFLOPS | Progress: (8/20) | 8.97 s
[Task 13/25] Current/Best: 6.18/ 18.41 GFLOPS | Progress: (12/20) | 11.86 s
[Task 13/25] Current/Best: 18.47/ 18.47 GFLOPS | Progress: (16/20) | 15.11 s
[Task 13/25] Current/Best: 14.46/ 18.47 GFLOPS | Progress: (20/20) | 17.80 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 15.21/ 15.21 GFLOPS | Progress: (4/20) | 5.89 s
[Task 14/25] Current/Best: 14.62/ 15.64 GFLOPS | Progress: (8/20) | 9.74 s
[Task 14/25] Current/Best: 17.64/ 17.64 GFLOPS | Progress: (12/20) | 11.98 s
[Task 14/25] Current/Best: 20.81/ 20.81 GFLOPS | Progress: (16/20) | 16.06 s
[Task 14/25] Current/Best: 12.65/ 20.81 GFLOPS | Progress: (20/20) | 19.38 s Done.
+
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 21.78/ 21.78 GFLOPS | Progress: (4/20) | 4.53 s
[Task 15/25] Current/Best: 14.27/ 21.78 GFLOPS | Progress: (8/20) | 8.59 s
[Task 15/25] Current/Best: 11.92/ 21.78 GFLOPS | Progress: (12/20) | 12.35 s
[Task 15/25] Current/Best: 13.94/ 21.78 GFLOPS | Progress: (16/20) | 15.31 s
[Task 15/25] Current/Best: 15.02/ 21.95 GFLOPS | Progress: (20/20) | 16.99 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 11.29/ 17.91 GFLOPS | Progress: (4/20) | 5.67 s
[Task 16/25] Current/Best: 17.17/ 17.91 GFLOPS | Progress: (8/20) | 7.81 s
[Task 16/25] Current/Best: 14.08/ 20.80 GFLOPS | Progress: (12/20) | 9.46 s
[Task 16/25] Current/Best: 15.40/ 20.80 GFLOPS | Progress: (16/20) | 11.87 s
[Task 16/25] Current/Best: 11.72/ 20.80 GFLOPS | Progress: (20/20)
| 14.32 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 10.12/ 18.76 GFLOPS | Progress: (4/20) | 5.56 s
[Task 17/25] Current/Best: 8.29/ 18.76 GFLOPS | Progress: (8/20) | 9.35 s
[Task 17/25] Current/Best: 12.86/ 21.30 GFLOPS | Progress: (12/20) | 11.98 s
[Task 17/25] Current/Best: 6.87/ 23.48 GFLOPS | Progress: (16/20) | 14.64 s
[Task 17/25] Current/Best: 7.51/ 23.48 GFLOPS | Progress: (20/20) | 17.52 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 15.61/ 17.51 GFLOPS | Progress: (4/20) | 5.85 s
[Task 18/25] Current/Best: 15.61/ 17.51 GFLOPS | Progress: (8/20) | 9.05 s
[Task 18/25] Current/Best: 15.24/ 18.78 GFLOPS | Progress: (12/20) | 10.92 s
[Task 18/25] Current/Best: 14.67/ 19.00 GFLOPS | Progress: (16/20) | 13.13 s
[Task 18/25] Current/Best: 12.13/ 19.00 GFLOPS | Progress: (20/20) | 15.63 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 11.52/ 17.32 GFLOPS | Progress: (4/20) | 7.07 s
[Task 19/25] Current/Best: 18.11/ 21.90 GFLOPS | Progress: (8/20) | 10.82 s
[Task 19/25] Current/Best: 20.81/ 21.90 GFLOPS | Progress: (12/20) | 14.80 s
[Task 19/25] Current/Best: 8.31/ 21.90 GFLOPS | Progress: (16/20) | 19.33 s
[Task 19/25] Current/Best: 18.55/ 21.90 GFLOPS | Progress: (20/20) | 22.87 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 14.49/ 14.49 GFLOPS | Progress: (4/20) | 4.57 s
[Task 20/25] Current/Best: 16.41/ 16.41 GFLOPS | Progress: (8/20) | 8.75 s
[Task 20/25] Current/Best: 7.96/ 16.41 GFLOPS | Progress: (12/20) | 11.95 s
[Task 20/25] Current/Best: 6.85/ 16.41 GFLOPS | Progress: (16/20) | 14.86 s
[Task 20/25] Current/Best: 15.98/ 20.51 GFLOPS | Progress: (20/20) | 18.67 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.95/ 22.24 GFLOPS | Progress: (4/20) | 6.16 s
[Task 21/25] Current/Best: 16.41/ 22.24 GFLOPS | Progress: (8/20) | 7.49 s Done.
+ Done.
+
[Task 21/25] Current/Best: 14.02/ 22.24 GFLOPS | Progress: (12/20) | 10.19 s
[Task 21/25] Current/Best: 1.62/ 22.24 GFLOPS | Progress: (16/20) | 13.58 s
[Task 21/25] Current/Best: 5.37/ 22.24 GFLOPS | Progress: (20/20) | 16.94 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 18.79/ 19.11 GFLOPS | Progress: (4/20) | 4.60 s
[Task 22/25] Current/Best: 6.74/ 19.11 GFLOPS | Progress: (8/20) | 7.04 s
[Task 22/25] Current/Best: 17.51/ 20.39 GFLOPS | Progress: (12/20) | 9.32 s
[Task 22/25] Current/Best: 17.81/ 20.39 GFLOPS | Progress: (16/20) | 11.69 s
[Task 22/25] Current/Best: 2.52/ 20.39 GFLOPS | Progress: (20/20) | 14.61 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 18.66/ 18.66 GFLOPS | Progress: (4/20) | 5.85 s
[Task 23/25] Current/Best: 13.04/ 18.66 GFLOPS | Progress: (8/20) | 9.81 s
[Task 23/25] Current/Best: 18.53/ 21.98 GFLOPS | Progress: (12/20) | 13.22 s
[Task 23/25] Current/Best: 10.96/ 21.98 GFLOPS | Progress: (16/20) | 17.11 s
[Task 23/25] Current/Best: 17.56/ 23.04 GFLOPS | Progress: (20/20) | 19.43 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 1.70/ 7.95 GFLOPS | Progress: (4/20) | 10.84 s
[Task 24/25] Current/Best: 6.93/ 7.95 GFLOPS | Progress: (8/20) | 21.87 s
[Task 24/25] Current/Best: 2.85/ 7.95 GFLOPS | Progress: (12/20) | 32.57 s
[Task 24/25] Current/Best: 4.47/ 7.95 GFLOPS | Progress: (16/20) | 45.19 s
[Task 24/25] Current/Best: 5.74/ 7.95 GFLOPS | Progress: (20/20) | 48.64 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 25/25] Current/Best: 3.66/ 7.61 GFLOPS | Progress: (4/20) | 13.78 s
[Task 25/25] Current/Best: 6.35/ 7.86 GFLOPS | Progress: (8/20) | 25.25 s
[Task 25/25] Current/Best: 7.52/ 9.56 GFLOPS | Progress: (12/20) | 27.52 s
[Task 25/25] Current/Best: 8.85/ 9.56 GFLOPS | Progress: (16/20) | 29.49 s
[Task 25/25] Current/Best: 1.54/ 9.56 GFLOPS | Progress: (20/20) | 40.46 s
@@ -721,8 +722,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 438.2907780400001, 'median': 438.296174549987, 'std': 2.0377520090315553}
- unoptimized: {'mean': 518.4311554900012, 'median': 519.1672235000055, 'std': 1.8314267201220487}
+ optimized: {'mean': 411.1730001000001, 'median': 410.0126122499887, 'std': 4.507568327493787}
+ unoptimized: {'mean': 515.2317357000004, 'median': 515.0500995000016, 'std': 2.6282614626980396}
@@ -745,7 +746,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 13 minutes 2.107 seconds)
+ **Total running time of the script:** ( 12 minutes 21.163 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 5c0f660453..e5eda65042 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
- 1.338e-07 secs/op
+ 1.42e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index d54acaf479..c5ca34092b 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, 0x225e56d0)), stage(b, placeholder(b, 0xe09bf80)), 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, 0x21b8c760)), stage(b, placeholder(b, 0x15dc8da0)), 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 [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index b3adf9f43a..f7a909468d 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,32 +5,32 @@
Computation times
=================
-**16:41.653** total execution time for **tutorial** files:
+**16:09.460** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 13:02.107 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 12:21.163 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:28.268 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:35.808 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:00.662 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.092 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:36.484 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:37.336 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:31.529 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:31.283 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.593 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.731 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.841 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.854 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.169 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.191 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``) | 00:00.000 | 0.0 MB |
++------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.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 |
-+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_install.py` (``install.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 150ef84ae4..289cb92787 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -285,8 +285,8 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000012
- naive: 0.000011
+ Numpy running time: 0.000007
+ naive: 0.000009
@@ -389,7 +389,7 @@ compile and run this new schedule with the parallel operation applied:
.. code-block:: none
- parallel: 0.000006
+ parallel: 0.000009
@@ -498,10 +498,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 1.2478570001803746e-05 1.0
- naive 1.11312e-05 0.8920252880250712
- parallel 6.0550000000000005e-06 0.48523188146756924
- vector 2.60456e-05 2.087226340537031
+ numpy 7.113379999736935e-06 1.0
+ naive 8.8491e-06 1.2440077713164845
+ parallel 9.0986e-06 1.279082517781488
+ vector 2.56191e-05 3.6015368222908712
@@ -922,7 +922,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.018123
+ Numpy running time: 0.019169
@@ -980,7 +980,7 @@ optimizations.
.. code-block:: none
- none: 3.388691
+ none: 3.396508
@@ -1080,7 +1080,7 @@ schedule.
.. code-block:: none
- blocking: 0.295139
+ blocking: 0.300891
@@ -1164,7 +1164,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.334693
+ vectorization: 0.346426
# 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: 0.117831
+ loop permutation: 0.119440
# 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: 0.109352
+ array packing: 0.107678
# 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: 0.111463
+ block caching: 0.110556
# 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: 0.146938
+ parallelization: 0.146284
# 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 3.388690665 1.0
- blocking 0.2951394141 0.08709541332537062
- vectorization 0.33469251460000005 0.09876750275758796
- loop permutation 0.11783072030000001 0.03477175462399428
- array packing 0.1093519739 0.03226968310487555
- block caching 0.11146290210000001 0.03289261638757458
- parallelization 0.1469383191 0.043361384565917586
+ none 3.3965078962 1.0
+ blocking 0.3008911487 0.08858838486335799
+ vectorization 0.3464257872 0.1019946950771349
+ loop permutation 0.1194403888 0.03516564437657556
+ array packing 0.10767792260000002 0.03170253857512584
+ block caching 0.11055616460000002 0.032549950707810756
+ parallelization 0.1462836534 0.04306883948765777
@@ -1596,7 +1596,7 @@ the computation for specific platforms.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 0.662 seconds)
+ **Total running time of the script:** ( 1 minutes 1.092 seconds)
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 9697735fef..f118b4b3b4 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-b228037a291bbc5d97fbeadceb75efc8fb94f2c7
+460374fed5f805a0ceb2b0a8de07c190ab90d83f
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index 534a0ae815..4f0cdc503d 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 21.056 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.136 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 06f44d77b0..337372eba2 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.zipd09201e7-5a6f-4f2d-8680-3686aa87f71c 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.zip2fdde9ab-eec0-4dc4-b7bb-73a56e975742 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 7cd290d9d8..a48bc55056 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]
- 19%|#8 | 7.88M/41.5M [00:00<00:00, 82.6MB/s]
- 38%|###7 | 15.8M/41.5M [00:00<00:00, 43.0MB/s]
- 54%|#####3 | 22.3M/41.5M [00:00<00:00, 42.6MB/s]
- 65%|######4 | 26.9M/41.5M [00:00<00:00, 33.1MB/s]
- 77%|#######7 | 32.0M/41.5M [00:00<00:00, 32.7MB/s]
- 85%|########5 | 35.4M/41.5M [00:01<00:00, 29.7MB/s]
- 92%|#########2| 38.4M/41.5M [00:01<00:00, 29.8MB/s]
-100%|##########| 41.5M/41.5M [00:01<00:00, 34.8MB/s]
+ 19%|#9 | 7.99M/41.5M [00:00<00:00, 68.4MB/s]
+ 35%|###4 | 14.5M/41.5M [00:00<00:00, 57.6MB/s]
+ 48%|####8 | 20.0M/41.5M [00:00<00:00, 50.2MB/s]
+ 60%|#####9 | 24.9M/41.5M [00:00<00:00, 44.6MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 52.3MB/s]
+ 96%|#########6| 40.0M/41.5M [00:00<00:00, 56.4MB/s]
+100%|##########| 41.5M/41.5M [00:00<00:00, 55.7MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 5282c79f3e..5e5f2cdde8 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -437,10 +437,11 @@ 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]
- 32%|###2 | 14.5M/44.7M [00:00<00:00, 151MB/s]
- 65%|######4 | 28.9M/44.7M [00:00<00:00, 116MB/s]
- 91%|######### | 40.4M/44.7M [00:00<00:00, 109MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 113MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 69.2MB/s]
+ 36%|###5 | 16.0M/44.7M [00:00<00:00, 71.3MB/s]
+ 68%|######7 | 30.3M/44.7M [00:00<00:00, 96.6MB/s]
+ 88%|########8 | 39.5M/44.7M [00:00<00:00, 79.8MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 67.3MB/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 3445483b18..41886eab08 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 33.607 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 33.505 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 3b25488f9b..9b962b2055 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>06:51.258</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>06:58.716</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>01:33.607</p></td>
+<td><p>01:33.505</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>01:21.056</p></td>
+<td><p>01:23.136</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>00:55.542</p></td>
+<td><p>00:57.393</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>00:38.350</p></td>
+<td><p>00:38.905</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>00:33.216</p></td>
+<td><p>00:33.107</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>00:30.013</p></td>
+<td><p>00:30.891</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>00:28.594</p></td>
+<td><p>00:28.334</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>00:25.198</p></td>
+<td><p>00:26.543</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>00:22.902</p></td>
+<td><p>00:24.133</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>00:02.781</p></td>
+<td><p>00:02.768</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 6cf6671ddf..52d0f1ceef 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -799,7 +799,7 @@ Top5 predictions:
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 2535.4974 2535.2080 2538.1975 2533.6781 1.3019
+ 2752.7754 2752.7170 2756.0316 2750.1003 1.7882
</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 9b9ea46e68..425409e70a 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
@@ -443,23 +443,27 @@ to run this tutorial with a real device over rpc.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
8192/102967424 [..............................] - ETA: 0s
- 3776512/102967424 [>.............................] - ETA: 1s
- 10575872/102967424 [==>...........................] - ETA: 0s
- 16769024/102967424 [===>..........................] - ETA: 0s
- 25157632/102967424 [======>.......................] - ETA: 0s
- 33546240/102967424 [========>.....................] - ETA: 0s
- 41934848/102967424 [===========>..................] - ETA: 0s
- 50503680/102967424 [=============>................] - ETA: 0s
- 59097088/102967424 [================>.............] - ETA: 0s
- 64618496/102967424 [=================>............] - ETA: 0s
- 68820992/102967424 [===================>..........] - ETA: 0s
- 74702848/102967424 [====================>.........] - ETA: 0s
- 79716352/102967424 [======================>.......] - ETA: 0s
- 84606976/102967424 [=======================>......] - ETA: 0s
- 87810048/102967424 [========================>.....] - ETA: 0s
+ 6635520/102967424 [>.............................] - ETA: 1s
+ 11354112/102967424 [==>...........................] - ETA: 1s
+ 16769024/102967424 [===>..........................] - ETA: 1s
+ 18964480/102967424 [====>.........................] - ETA: 1s
+ 22216704/102967424 [=====>........................] - ETA: 1s
+ 25157632/102967424 [======>.......................] - ETA: 1s
+ 33259520/102967424 [========>.....................] - ETA: 1s
+ 33546240/102967424 [========>.....................] - ETA: 1s
+ 38854656/102967424 [==========>...................] - ETA: 1s
+ 41934848/102967424 [===========>..................] - ETA: 1s
+ 50323456/102967424 [=============>................] - ETA: 1s
+ 58712064/102967424 [================>.............] - ETA: 0s
+ 67100672/102967424 [==================>...........] - ETA: 0s
+ 69296128/102967424 [===================>..........] - ETA: 0s
+ 76079104/102967424 [=====================>........] - ETA: 0s
+ 83877888/102967424 [=======================>......] - ETA: 0s
+ 90718208/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
- 99999744/102967424 [============================>.] - ETA: 0s
-102967424/102967424 [==============================] - 1s 0us/step
+100646912/102967424 [============================>.] - ETA: 0s
+102850560/102967424 [============================>.] - ETA: 0s
+102967424/102967424 [==============================] - 2s 0us/step
</pre></div>
</div>
</div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 413fbccdc5..52125fc380 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)
- 16.1580 16.0368 16.8793 15.9033 0.2722
+ 16.6141 16.7317 16.8329 16.1908 0.2325
</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 cb403df3e7..38d95b3ceb 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -459,28 +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<00:02, 58.3MB/s]
- 9%|9 | 16.0M/170M [00:00<00:02, 76.1MB/s]
- 17%|#6 | 28.2M/170M [00:00<00:01, 87.6MB/s]
- 21%|##1 | 36.5M/170M [00:00<00:01, 85.9MB/s]
- 27%|##6 | 45.1M/170M [00:00<00:01, 87.5MB/s]
- 31%|###1 | 53.5M/170M [00:00<00:01, 78.2MB/s]
- 37%|###7 | 63.7M/170M [00:00<00:01, 86.6MB/s]
- 42%|####2 | 72.1M/170M [00:00<00:01, 78.0MB/s]
- 47%|####7 | 80.0M/170M [00:01<00:01, 71.9MB/s]
- 51%|#####1 | 87.1M/170M [00:01<00:01, 71.8MB/s]
- 55%|#####5 | 94.0M/170M [00:01<00:01, 66.7MB/s]
- 60%|###### | 102M/170M [00:01<00:01, 68.6MB/s]
- 64%|######4 | 109M/170M [00:01<00:00, 63.9MB/s]
- 68%|######7 | 115M/170M [00:01<00:01, 50.6MB/s]
- 71%|####### | 120M/170M [00:01<00:01, 39.4MB/s]
- 74%|#######3 | 125M/170M [00:02<00:01, 39.5MB/s]
- 79%|#######9 | 134M/170M [00:02<00:00, 46.6MB/s]
- 82%|########1 | 139M/170M [00:02<00:00, 46.4MB/s]
- 88%|########8 | 150M/170M [00:02<00:00, 61.1MB/s]
- 93%|#########3| 158M/170M [00:02<00:00, 60.4MB/s]
- 97%|#########6| 164M/170M [00:02<00:00, 59.0MB/s]
-100%|##########| 170M/170M [00:02<00:00, 60.9MB/s]
+ 4%|3 | 6.30M/170M [00:00<00:02, 61.7MB/s]
+ 7%|7 | 12.2M/170M [00:00<00:03, 54.5MB/s]
+ 10%|# | 17.4M/170M [00:00<00:03, 40.9MB/s]
+ 14%|#4 | 24.0M/170M [00:00<00:03, 45.1MB/s]
+ 19%|#8 | 32.0M/170M [00:00<00:02, 53.2MB/s]
+ 24%|##3 | 40.0M/170M [00:00<00:02, 52.4MB/s]
+ 28%|##8 | 48.0M/170M [00:00<00:02, 58.9MB/s]
+ 33%|###2 | 56.0M/170M [00:01<00:01, 61.4MB/s]
+ 38%|###7 | 64.0M/170M [00:01<00:01, 60.4MB/s]
+ 42%|####2 | 72.0M/170M [00:01<00:01, 62.1MB/s]
+ 46%|####6 | 78.3M/170M [00:01<00:01, 58.5MB/s]
+ 50%|##### | 85.7M/170M [00:01<00:01, 63.3MB/s]
+ 54%|#####4 | 91.9M/170M [00:01<00:01, 51.0MB/s]
+ 57%|#####7 | 97.2M/170M [00:01<00:01, 43.1MB/s]
+ 60%|###### | 102M/170M [00:02<00:01, 42.1MB/s]
+ 63%|######2 | 107M/170M [00:02<00:01, 37.0MB/s]
+ 66%|######5 | 112M/170M [00:02<00:01, 40.2MB/s]
+ 71%|####### | 120M/170M [00:02<00:01, 48.4MB/s]
+ 75%|#######5 | 128M/170M [00:02<00:00, 48.5MB/s]
+ 79%|#######9 | 134M/170M [00:02<00:00, 50.3MB/s]
+ 82%|########2 | 139M/170M [00:02<00:00, 43.8MB/s]
+ 85%|########4 | 144M/170M [00:03<00:00, 33.4MB/s]
+ 89%|########9 | 152M/170M [00:03<00:00, 36.1MB/s]
+ 93%|#########3| 158M/170M [00:03<00:00, 39.3MB/s]
+ 96%|#########5| 162M/170M [00:03<00:00, 35.6MB/s]
+ 98%|#########7| 166M/170M [00:03<00:00, 36.2MB/s]
+100%|##########| 170M/170M [00:03<00:00, 46.0MB/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=& [...]
@@ -578,7 +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 32.589 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 40.667 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 f15d0de71a..60bfa20483 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -500,8 +500,9 @@ training. Other models require a full post training calibration.</p>
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, 79.7MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 98.8MB/s]
+ 59%|#####8 | 7.99M/13.6M [00:00<00:00, 41.0MB/s]
+ 90%|########9 | 12.2M/13.6M [00:00<00:00, 38.3MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 42.4MB/s]
</pre></div>
</div>
</div>
@@ -592,7 +593,7 @@ 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)
- 90.2261 90.1194 95.6162 89.8526 0.6157
+ 90.5014 90.3875 95.4684 90.0507 0.6352
</pre></div>
</div>
<div class="admonition note">
@@ -631,7 +632,7 @@ 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 16.221 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 17.885 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 0d8873564d..6f1f38adeb 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)
- 119.3129 119.4781 122.7859 117.0400 0.9863
+ 120.8390 120.7592 123.9352 120.0472 0.4423
</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 35.781 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 29.508 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 0b9142689b..727f561a5b 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 31.367 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 44.272 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 a07b747d40..1f49b717da 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -468,24 +468,25 @@ 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]
- 1%| | 1202/132723 [00:00<00:10, 11980.25KB/s]
- 3%|2 | 3618/132723 [00:00<00:06, 19104.97KB/s]
- 6%|6 | 8612/132723 [00:00<00:03, 33160.88KB/s]
- 13%|#2 | 17108/132723 [00:00<00:02, 53587.86KB/s]
- 20%|#9 | 25895/132723 [00:00<00:01, 65941.20KB/s]
- 26%|##6 | 34731/132723 [00:00<00:01, 73557.54KB/s]
- 33%|###2 | 43645/132723 [00:00<00:01, 78649.02KB/s]
- 40%|###9 | 52464/132723 [00:00<00:00, 81683.19KB/s]
- 46%|####6 | 61317/132723 [00:00<00:00, 83821.06KB/s]
- 53%|#####2 | 70217/132723 [00:01<00:00, 85417.38KB/s]
- 60%|#####9 | 79058/132723 [00:01<00:00, 86330.36KB/s]
- 66%|######6 | 87898/132723 [00:01<00:00, 86958.22KB/s]
- 73%|#######2 | 96818/132723 [00:01<00:00, 87635.34KB/s]
- 80%|#######9 | 105721/132723 [00:01<00:00, 88054.31KB/s]
- 86%|########6 | 114706/132723 [00:01<00:00, 88590.67KB/s]
- 93%|#########3| 123566/132723 [00:01<00:00, 88586.83KB/s]
-100%|#########9| 132503/132723 [00:01<00:00, 88821.31KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 77819.97KB/s]
+ 1%| | 1220/132723 [00:00<00:10, 12146.18KB/s]
+ 3%|2 | 3867/132723 [00:00<00:06, 20556.05KB/s]
+ 7%|7 | 9608/132723 [00:00<00:03, 37364.91KB/s]
+ 13%|#3 | 17405/132723 [00:00<00:02, 53381.44KB/s]
+ 19%|#9 | 25323/132723 [00:00<00:01, 62679.39KB/s]
+ 25%|##4 | 33138/132723 [00:00<00:01, 67935.15KB/s]
+ 31%|### | 40718/132723 [00:00<00:01, 70502.75KB/s]
+ 37%|###6 | 48701/132723 [00:00<00:01, 73467.38KB/s]
+ 43%|####2 | 56662/132723 [00:00<00:01, 75385.78KB/s]
+ 49%|####8 | 64535/132723 [00:01<00:00, 76415.06KB/s]
+ 55%|#####4 | 72474/132723 [00:01<00:00, 77323.26KB/s]
+ 61%|###### | 80412/132723 [00:01<00:00, 77945.28KB/s]
+ 67%|######6 | 88370/132723 [00:01<00:00, 78439.05KB/s]
+ 73%|#######2 | 96329/132723 [00:01<00:00, 78784.59KB/s]
+ 79%|#######8 | 104266/132723 [00:01<00:00, 78960.04KB/s]
+ 85%|########4 | 112219/132723 [00:01<00:00, 79129.96KB/s]
+ 91%|######### | 120226/132723 [00:01<00:00, 79411.10KB/s]
+ 97%|#########6| 128205/132723 [00:01<00:00, 79523.49KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 71392.30KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -524,7 +525,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 53.392 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 1.809 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 d943befec2..fcd18b8a4d 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>16:15.107</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>16:45.389</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -354,43 +354,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>03:53.392</p></td>
+<td><p>04:01.809</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>03:32.589</p></td>
+<td><p>03:40.667</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>02:35.781</p></td>
+<td><p>02:29.508</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>01:31.367</p></td>
+<td><p>01:44.272</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>01:16.221</p></td>
+<td><p>01:17.885</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>00:54.735</p></td>
+<td><p>00:58.168</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>00:50.517</p></td>
+<td><p>00:52.682</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>00:42.972</p></td>
+<td><p>00:43.631</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:28.910</p></td>
+<td><p>00:28.632</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:28.617</p></td>
+<td><p>00:28.129</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 059455be17..6ca0f31ace 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.zip6506fbaf-67a8-4e5d-a4c9-8a33d6ba9b1a 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.zip1960c2b8-74ce-4e50-8d2e-c7a2a7cf59c6 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 1494055fb0..e842315ab7 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>00:54.069</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:57.026</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:50.289</p></td>
+<td><p>00:53.068</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>00:02.714</p></td>
+<td><p>00:02.838</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>00:01.059</p></td>
+<td><p>00:01.113</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 491a0f5885..5e5bc71458 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: 22136us [22136us] (48.53%; 48.53%)
-FoldScaleAxis: 23475us [6us] (51.47%; 51.47%)
- FoldConstant: 23469us [1682us] (51.45%; 99.97%)
- InferType: 21786us [21786us] (47.77%; 92.83%)
+InferType: 22858us [22858us] (48.70%; 48.70%)
+FoldScaleAxis: 24075us [8us] (51.30%; 51.30%)
+ FoldConstant: 24067us [1807us] (51.28%; 99.97%)
+ InferType: 22260us [22260us] (47.43%; 92.49%)
</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: 21436us [21436us] (48.12%; 48.12%)
-FoldScaleAxis: 23112us [5us] (51.88%; 51.88%)
- FoldConstant: 23107us [1703us] (51.87%; 99.98%)
- InferType: 21404us [21404us] (48.05%; 92.63%)
+InferType: 22459us [22459us] (48.62%; 48.62%)
+FoldScaleAxis: 23732us [8us] (51.38%; 51.38%)
+ FoldConstant: 23724us [1810us] (51.36%; 99.96%)
+ InferType: 21914us [21914us] (47.44%; 92.37%)
</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 7f36f41f97..58446b678f 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: 45.412864 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 53.518367 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 6e258fbe73..ed40e6b2ed 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: 11.515900 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 12.243971 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 7849eb48e2..b9395bc799 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: 0.017848
-Baseline: 3.374923
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018984
+Baseline: 3.358529
</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: 0.297316
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.308085
</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: 0.335475
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.347758
</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: 0.118474
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116191
</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: 0.109288
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.108726
</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: 0.111118
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.110817
</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: 0.146157
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.148391
</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 d34967ed34..d5acfbd2a9 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:35.195</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.585</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:32.188</p></td>
+<td><p>00:32.595</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>00:01.847</p></td>
+<td><p>00:01.904</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></td>
-<td><p>00:01.160</p></td>
+<td><p>00:01.087</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 f744ac56e5..41fdff014d 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>09:58.930</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>10:15.867</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>06:05.440</p></td>
+<td><p>06:18.394</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>01:42.146</p></td>
+<td><p>01:44.131</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>01:11.032</p></td>
+<td><p>01:12.362</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></td>
-<td><p>00:32.895</p></td>
+<td><p>00:32.469</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>00:13.997</p></td>
+<td><p>00:14.549</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:13.420</p></td>
+<td><p>00:13.962</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 2e68ee73ed..8c55d77450 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
@@ -511,107 +511,869 @@ class Module:
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
blockIdx_x = T.launch_thread("blockIdx.x", 32)
- conv2d_nchw = T.allocate([4], "float32", "local")
- pad_temp_shared = T.allocate([162], "float32", "shared")
- kernel_shared = T.allocate([288], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 196)
- conv2d_nchw_1 = T.Buffer((4,), data=conv2d_nchw, scope="local", align=16)
+ conv2d_nchw = T.allocate([28], "float32", "local")
+ pad_temp_shared = T.allocate([2592], "float32", "shared")
+ kernel_shared = T.allocate([4608], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 28)
+ conv2d_nchw_1 = T.Buffer((196,), data=conv2d_nchw, scope="local", align=32)
conv2d_nchw_1[0] = T.float32(0)
+ conv2d_nchw_1[14] = T.float32(0)
+ conv2d_nchw_1[7] = T.float32(0)
+ conv2d_nchw_1[21] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
+ conv2d_nchw_1[15] = T.float32(0)
+ conv2d_nchw_1[8] = T.float32(0)
+ conv2d_nchw_1[22] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
+ conv2d_nchw_1[16] = T.float32(0)
+ conv2d_nchw_1[9] = T.float32(0)
+ conv2d_nchw_1[23] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
- for rc_outer_outer in range(256):
- pad_temp_shared_1 = T.Buffer((162,), data=pad_temp_shared, scope="shared")
- with T.launch_thread("threadIdx.x", 196) as threadIdx_x_1:
- if T.likely(threadIdx_x_1 < 54):
- data_1 = T.Buffer((25088,), data=data.data)
- pad_temp_shared_1[threadIdx_x_1 * 3] = T.if_then_else(3 <= threadIdx_x_1 % 27 and threadIdx_x_1 % 27 < 24 and 1 <= threadIdx_x_1 % 3, data_1[rc_outer_outer * 98 + threadIdx_x_1 // 27 * 49 + threadIdx_x_1 % 27 // 3 * 7 + threadIdx_x_1 % 3 * 3 - 8], T.float32(0))
- pad_temp_shared_1[threadIdx_x_1 * 3 + 1] = T.if_then_else(3 <= threadIdx_x_1 % 27 and threadIdx_x_1 % 27 < 24, data_1[rc_outer_outer * 98 + threadIdx_x_1 // 27 * 49 + threadIdx_x_1 % 27 // 3 * 7 + threadIdx_x_1 % 3 * 3 - 7], T.float32(0))
- pad_temp_shared_1[threadIdx_x_1 * 3 + 2] = T.if_then_else(3 <= threadIdx_x_1 % 27 and threadIdx_x_1 % 27 < 24 and threadIdx_x_1 % 3 < 2, data_1[rc_outer_outer * 98 + threadIdx_x_1 // 27 * 49 + threadIdx_x_1 % 27 // 3 * 7 + threadIdx_x_1 % 3 * 3 - 6], T.float32(0))
+ conv2d_nchw_1[17] = T.float32(0)
+ conv2d_nchw_1[10] = T.float32(0)
+ conv2d_nchw_1[24] = T.float32(0)
+ conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[18] = T.float32(0)
+ conv2d_nchw_1[11] = T.float32(0)
+ conv2d_nchw_1[25] = T.float32(0)
+ conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[19] = T.float32(0)
+ conv2d_nchw_1[12] = T.float32(0)
+ conv2d_nchw_1[26] = T.float32(0)
+ conv2d_nchw_1[6] = T.float32(0)
+ conv2d_nchw_1[20] = T.float32(0)
+ conv2d_nchw_1[13] = T.float32(0)
+ conv2d_nchw_1[27] = T.float32(0)
+ for rc_outer_outer in range(16):
+ cse_var_2: T.int32 = rc_outer_outer * 1568
+ cse_var_1: T.int32 = rc_outer_outer * 288
threadIdx_x_1 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((288,), data=kernel_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((2592,), data=pad_temp_shared, scope="shared")
+ data_1 = T.Buffer((25088,), data=data.data)
+ with T.launch_thread(threadIdx_x_1, 28):
+ pad_temp_shared_1[threadIdx_x_1 * 36] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 3] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 4] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 5] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 6] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 7] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 8] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 9] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 10] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 11] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 12] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 13] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 14] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 15] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 16] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 17] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 18] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 19] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 20] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 21] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 22] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 23] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 24] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 25] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 26] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 27] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 28] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 29] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 30] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 31] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 32] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 33] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 34] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + (threadIdx_x_1 * 4 + 3) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 35] = T.float32(0)
+ with T.launch_thread(threadIdx_x_1, 28):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1008] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1009] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1010] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1011] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1012] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1013] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1014] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1015] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 4) % 9 and (threadIdx_x_1 * 4 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 112) // 9 * 49 + (threadIdx_x_1 * 4 + 4) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1016] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1017] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1018] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1019] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1020] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1021] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1022] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1023] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1024] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 5) % 9 and (threadIdx_x_1 * 4 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 113) // 9 * 49 + (threadIdx_x_1 * 4 + 5) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1025] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1026] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1027] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1028] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1029] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1030] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1031] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1032] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1033] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 6) % 9 and (threadIdx_x_1 * 4 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 114) // 9 * 49 + (threadIdx_x_1 * 4 + 6) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1034] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1035] = T.float32(0)
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1036] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 7], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1037] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 6], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1038] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 5], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1039] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 4], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1040] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 3], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1041] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 2], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1042] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 7) % 9 and (threadIdx_x_1 * 4 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 115) // 9 * 49 + (threadIdx_x_1 * 4 + 7) % 9 * 7 - 1], T.float32(0))
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 1043] = T.float32(0)
+ with T.launch_thread(threadIdx_x_1, 28):
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2016] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2017] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 7], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2018] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 6], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2019] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 5], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2020] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 4], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2021] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 3], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2022] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 2], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2023] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 8) % 9 and (threadIdx_x_1 * 4 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 224) // 9 * 49 + (threadIdx_x_1 * 4 + 8) % 9 * 7 - 1], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2024] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2025] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2026] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1218], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2027] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1219], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2028] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1220], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2029] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1221], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2030] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1222], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2031] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1223], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2032] = T.if_then_else(1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 * 4 // 9 * 49 + threadIdx_x_1 * 4 % 9 * 7 + 1224], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2033] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2034] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2035] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 7], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2036] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 6], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2037] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 5], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2038] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 4], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2039] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 3], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2040] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 2], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2041] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 226) // 9 * 49 + (threadIdx_x_1 * 4 + 1) % 9 * 7 - 1], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2042] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2043] = T.float32(0)
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2044] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 7], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2045] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 6], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2046] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 5], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2047] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 4], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2048] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 3], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2049] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 2], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2050] = T.if_then_else(1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 * 4 + 227) // 9 * 49 + (threadIdx_x_1 * 4 + 2) % 9 * 7 - 1], T.float32(0))
+ if T.likely(threadIdx_x_1 < 16):
+ pad_temp_shared_1[threadIdx_x_1 * 36 + 2051] = T.float32(0)
+ threadIdx_x_2 = T.env_thread("threadIdx.x")
+ kernel_shared_1 = T.Buffer((4608,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_1, 196):
- kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x * 73728 + threadIdx_x_1 // 18 * 4608 + rc_outer_outer * 18 + threadIdx_x_1 % 18]
- with T.launch_thread(threadIdx_x_1, 196):
- if T.likely(threadIdx_x_1 < 92):
- kernel_shared_1[threadIdx_x_1 + 196] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 196) // 18 * 4608 + rc_outer_outer * 18 + (threadIdx_x_1 + 16) % 18 // 3 * 3 + (threadIdx_x_1 + 1) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72 + 36]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 72 + 54]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 37]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 72 + 55]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 38]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 72 + 56]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 39]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 72 + 57]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 40]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 72 + 58]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 41]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 72 + 59]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 24]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 42]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 72 + 60]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 25]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 43]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 72 + 61]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 26]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 44]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 72 + 62]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 27]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 45]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 72 + 63]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 28]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 46]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 72 + 64]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 29]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 47]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 72 + 65]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 30]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 72 + 66]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 31]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 49]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 72 + 67]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 32]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 50]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 72 + 68]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 33]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 51]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 72 + 69]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 34]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 52]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 72 + 70]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 35]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 53]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 72 + 71]
- for i1_inner in range(4):
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 28] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 56] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 56]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 84] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 112] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 112]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 140] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 168] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 196] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 196]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 224] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 224]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 252] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 252]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 280] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 280) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 308] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 308) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 20]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 336] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 336) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 364] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 364) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 392] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 392) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 420] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 420) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 448] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 448) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 160]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 476] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 476) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 504] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 504) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 216]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 532] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 532) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 244]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 560] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 560) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 272) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 588] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 588) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 12]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 616] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 616) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 40]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 644] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 644) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 68]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 672] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 672) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 96]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 700] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 700) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 124]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 728] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 728) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 152]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 756] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 756) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 180]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 784] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 784) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 208]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 812] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 812) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 236]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 840] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 840) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 264) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 868] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 868) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 4]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 896] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 896) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 32]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 924] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 924) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 60]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 952] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 952) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 88]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 980] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 980) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 116]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1008] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1008) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 144]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1036] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1036) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 172]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1064] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1064) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 200]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1092] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1092) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 228]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1120] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1120) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 256]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1148] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1148) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 284) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1176] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1176) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 24]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1204] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1204) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 52]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1232] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1232) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 80]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1260] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1260) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 108]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1288] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1288) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 136]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1316] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1316) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 164]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1344) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 192]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1372] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1372) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 220]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1400] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1400) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 248]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1428] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1428) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 276) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1456] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1456) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 16]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1484] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1484) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 44]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1512] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1512) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 72]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1540] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1540) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 100]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1568) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 128]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1596] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1596) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 156]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1624] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1624) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 184]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1652] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1652) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 212]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1680] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1680) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 240]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1708] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1708) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 268) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1736] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1736) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 8]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1764] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1764) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 36]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1792] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1792) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 64]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1820] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1820) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 92]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1848] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1848) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 120]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1876] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1876) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 148]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1904] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1904) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 176]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1932] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1932) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 204]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1960] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1960) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 232]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 1988] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 1988) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 260]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2016] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 32256]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2044] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2044) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2072] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2072) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 56]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2100] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2100) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2128] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2128) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 112]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2156] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2156) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2184] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2184) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2212] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2212) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 196]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2240] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2240) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 224]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2268] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2268) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 252]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2296] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2296) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2324] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2324) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 20]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2352] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2352) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2380] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2380) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2408] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2408) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2436] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2436) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2464] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2464) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 160]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2492] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2492) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2520] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2520) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 216]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2548] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2548) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 244]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2576] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2576) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 272) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2604] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2604) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 12]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2632] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2632) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 40]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2660] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2660) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 68]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2688) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 96]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2716] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2716) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 124]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2744] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2744) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 152]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2772] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2772) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 180]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2800] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2800) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 208]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2828] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2828) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 236]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2856] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2856) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 264) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2884] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2884) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 4]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2912] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2912) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 32]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2940] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2940) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 60]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2968] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2968) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 88]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 2996] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 2996) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 116]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3024] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3024) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 144]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3052] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3052) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 172]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3080] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3080) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 200]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3108] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3108) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 228]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3136] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3136) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 256]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3164] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3164) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 284) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3192] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3192) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 24]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3220] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3220) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 52]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3248] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3248) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 80]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3276] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3276) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 108]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3304] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3304) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 136]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3332] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3332) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 164]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3360] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3360) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 192]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3388] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3388) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 220]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3416] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3416) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 248]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3444] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3444) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 276) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3472] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3472) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 16]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3500] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3500) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 44]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3528] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3528) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 72]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3556] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3556) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 100]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3584] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3584) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 128]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3612] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3612) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 156]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3640] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3640) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 184]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3668] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3668) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 212]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3696] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3696) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 240]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3724] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3724) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 268) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3752] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3752) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 8]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3780] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3780) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 36]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3808] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3808) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 64]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3836] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3836) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 92]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3864] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3864) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 120]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3892] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3892) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 148]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3920] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3920) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 176]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3948] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3948) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 204]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 3976] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 3976) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 232]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4004] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4004) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 260]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4032] = kernel_1[blockIdx_x * 73728 + cse_var_1 + threadIdx_x_2 + 64512]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4060] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4060) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4088] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4088) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 56]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4116] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4116) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4144] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4144) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 112]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4172] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4172) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4200] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4200) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4228] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4228) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 196]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4256] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4256) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 224]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4284] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4284) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 252]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4312] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4312) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4340] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4340) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 20]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4368] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4368) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4396] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4396) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4424] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4424) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4452] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4452) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4480] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4480) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 160]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4508] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4508) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4536] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4536) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 216]
+ with T.launch_thread(threadIdx_x_2, 28):
+ kernel_shared_1[threadIdx_x_2 + 4564] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4564) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 244]
+ with T.launch_thread(threadIdx_x_2, 28):
+ if T.likely(threadIdx_x_2 < 16):
+ kernel_shared_1[threadIdx_x_2 + 4592] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 + 4592) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 272]
+ for rc_outer_inner, ry_outer_inner in T.grid(8, 3):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 171] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 252] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 180] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 261] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 27] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 108] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 189] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 270] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 36] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 117] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 198] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 279] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 45] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 126] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 207] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 288] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2304]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 288]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2592]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 9]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2313]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 297]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 135] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2601]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 18]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2322]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 306]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 216] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2610]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 27]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2331]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 315]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 297] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2619]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 172] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 253] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 181] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 262] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 28] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 109] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 190] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 271] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 37] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 118] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 199] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 280] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 46] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 127] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 208] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 289] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 1]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2305]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 289]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2593]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 10]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2314]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 298]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 136] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2602]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 19]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2323]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 307]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 217] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2611]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 28]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2332]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 316]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 298] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2620]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 173] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 254] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 182] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 263] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 29] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 110] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 191] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 272] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 38] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 119] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 200] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 281] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 47] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 128] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 209] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 290] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2306]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 290]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 56] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2594]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 11]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2315]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 299]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 137] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2603]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 20]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2324]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 308]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 218] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2612]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 29]
+ conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2333]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 317]
+ conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[rc_outer_inner * 324 + ry_outer_inner * 9 + threadIdx_x % 7 + 299] * kernel_shared_1[threadIdx_x // 7 * 576 + rc_outer_inner * 36 + ry_outer_inner * 3 + 2621]
+ for i1_inner, i2_inner in T.grid(2, 7):
+ cse_var_3: T.int32 = i1_inner * 7 + i2_inner
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x * 784 + threadIdx_x // 49 * 196 + i1_inner * 49 + threadIdx_x % 49] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 * 4 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x // 7 * 98 + i1_inner * 49 + i2_inner * 7 + threadIdx_x % 7] = T.max(conv2d_nchw_1[cse_var_3] + bias_1[blockIdx_x * 16 + threadIdx_x // 7 * 2 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x // 7 * 98 + i1_inner * 49 + i2_inner * 7 + threadIdx_x % 7 + 392] = T.max(conv2d_nchw_1[cse_var_3 + 14] + bias_1[blockIdx_x * 16 + threadIdx_x // 7 * 2 + i1_inner + 8], T.float32(0))
</pre></div>
</div>
</div>
@@ -645,7 +1407,7 @@ 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: 0.302 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.309 ms
</pre></div>
</div>
</div>
@@ -674,33 +1436,33 @@ conv2d_nchw_nn_o_i, conv2d_nchw_nn_i = s[conv2d_nchw].split(conv2d_nchw_nn, fact
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=4)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=4)
-conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
+conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
-conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
-conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
+conv2d_nchw_yy_o_o_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=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=1)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=8)
conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
-conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
-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_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
+compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=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=1)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
@@ -723,14 +1485,14 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=28)
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=3)
+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=36)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=28)
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", 1024)
+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)
CUDA source code:
@@ -755,101 +1517,734 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[4];
- __shared__ float pad_temp_shared[162];
- __shared__ float kernel_shared[288];
+extern "C" __global__ void __launch_bounds__(28) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[28];
+ __shared__ float pad_temp_shared[2592];
+ __shared__ float kernel_shared[4608];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[21] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[22] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[16] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[23] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 256; ++rc_outer_outer) {
+ conv2d_nchw[17] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[24] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[18] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[25] = 0.000000e+00f;
+ conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[19] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[26] = 0.000000e+00f;
+ conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[20] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[27] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
__syncthreads();
- if (((int)threadIdx.x) < 54) {
- pad_temp_shared[(((int)threadIdx.x) * 3)] = ((((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) && (1 <= (((int)threadIdx.x) % 3))) ? data[(((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + ((((int)threadIdx.x) % 3) * 3)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 3) + 1)] = (((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) ? data[(((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + ((((int)threadIdx.x) % 3) * 3)) - 7)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 3) + 2)] = ((((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) && ((((int)threadIdx.x) % 3) < 2)) ? data[(((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + ((((int)threadIdx.x) % 3) * 3)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) * 36)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 3)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 4)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 5)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 6)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 7)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 8)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 9)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 10)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 11)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 12)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 13)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 14)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 15)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 16)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 17)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 18)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 19)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 20)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 21)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 22)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 23)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 24)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 25)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 26)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 27)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 28)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 29)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 30)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 31)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 32)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 33)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 34)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 35)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1008)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1009)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1010)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1011)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1012)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1013)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1014)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1015)] = (((1 <= (((((int)threadIdx.x) * 4) + 4) % 9)) && ((((((int)threadIdx.x) * 4) + 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 112) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 4) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1016)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1017)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1018)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1019)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1020)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1021)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1022)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1023)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1024)] = (((1 <= (((((int)threadIdx.x) * 4) + 5) % 9)) && ((((((int)threadIdx.x) * 4) + 5) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 113) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 5) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1025)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1026)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1027)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1028)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1029)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1030)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1031)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1032)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1033)] = (((1 <= (((((int)threadIdx.x) * 4) + 6) % 9)) && ((((((int)threadIdx.x) * 4) + 6) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 114) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 6) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1034)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1035)] = 0.000000e+00f;
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1036)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1037)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1038)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1039)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 4)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1040)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 3)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1041)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 2)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1042)] = (((1 <= (((((int)threadIdx.x) * 4) + 7) % 9)) && ((((((int)threadIdx.x) * 4) + 7) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 115) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 7) % 9) * 7)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 1043)] = 0.000000e+00f;
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2016)] = 0.000000e+00f;
}
- kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 18)) + (((int)threadIdx.x) % 18))];
- if (((int)threadIdx.x) < 92) {
- kernel_shared[(((int)threadIdx.x) + 196)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 196) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((((int)threadIdx.x) + 16) % 18) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2017)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 7)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2018)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 6)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2019)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 5)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2020)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 4)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2021)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 3)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2022)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 2)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2023)] = (((1 <= (((((int)threadIdx.x) * 4) + 8) % 9)) && ((((((int)threadIdx.x) * 4) + 8) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 224) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 8) % 9) * 7)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2024)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2025)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2026)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1218)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2027)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1219)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2028)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1220)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2029)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1221)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2030)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1222)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2031)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1223)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2032)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) + 1224)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2033)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2034)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2035)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 7)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2036)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 6)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2037)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 5)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2038)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 4)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2039)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 3)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2040)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 2)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2041)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 226) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2042)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2043)] = 0.000000e+00f;
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2044)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 7)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2045)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 6)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2046)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 5)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2047)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 4)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2048)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 3)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2049)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 2)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2050)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 227) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 16) {
+ pad_temp_shared[((((int)threadIdx.x) * 36) + 2051)] = 0.000000e+00f;
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 28)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 56)];
+ kernel_shared[(((int)threadIdx.x) + 84)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 112)];
+ kernel_shared[(((int)threadIdx.x) + 140)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 196)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 224)];
+ kernel_shared[(((int)threadIdx.x) + 252)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 252)];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 280) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 308)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 308) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 20)];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 336) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 364)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 364) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 392) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 420)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 420) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 448) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 160)];
+ kernel_shared[(((int)threadIdx.x) + 476)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 476) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 504) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 532)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 532) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 244)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 560) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 272) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 588) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 12)];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 616) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 40)];
+ kernel_shared[(((int)threadIdx.x) + 644)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 644) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 68)];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 672) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 700)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 700) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 124)];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 728) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 152)];
+ kernel_shared[(((int)threadIdx.x) + 756)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 756) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 180)];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 208)];
+ kernel_shared[(((int)threadIdx.x) + 812)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 812) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 236)];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 840) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 264) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 868)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 868) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 4)];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 896) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 32)];
+ kernel_shared[(((int)threadIdx.x) + 924)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 924) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 60)];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 952) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 88)];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 980) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 116)];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1008) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 1036)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1036) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 172)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1064) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 200)];
+ kernel_shared[(((int)threadIdx.x) + 1092)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1092) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 228)];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1120) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 256)];
+ kernel_shared[(((int)threadIdx.x) + 1148)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1148) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 284) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1176) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 1204)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1204) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 52)];
+ kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1232) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 80)];
+ kernel_shared[(((int)threadIdx.x) + 1260)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1260) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 108)];
+ kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1288) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 136)];
+ kernel_shared[(((int)threadIdx.x) + 1316)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1316) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 164)];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1344) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 192)];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1372) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 220)];
+ kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1400) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 248)];
+ kernel_shared[(((int)threadIdx.x) + 1428)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1428) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 276) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1456) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 16)];
+ kernel_shared[(((int)threadIdx.x) + 1484)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1484) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 44)];
+ kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1512) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 1540)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1540) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 100)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1568) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 128)];
+ kernel_shared[(((int)threadIdx.x) + 1596)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1596) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 156)];
+ kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1624) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 184)];
+ kernel_shared[(((int)threadIdx.x) + 1652)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1652) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 212)];
+ kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1680) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 240)];
+ kernel_shared[(((int)threadIdx.x) + 1708)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1708) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 268) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1736) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 8)];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1764) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 36)];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1792) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64)];
+ kernel_shared[(((int)threadIdx.x) + 1820)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1820) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 92)];
+ kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1848) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 120)];
+ kernel_shared[(((int)threadIdx.x) + 1876)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1876) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 148)];
+ kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1904) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 176)];
+ kernel_shared[(((int)threadIdx.x) + 1932)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1932) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 204)];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1960) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 232)];
+ kernel_shared[(((int)threadIdx.x) + 1988)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1988) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 260)];
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 2044)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2044) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 2072)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2072) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 56)];
+ kernel_shared[(((int)threadIdx.x) + 2100)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2100) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2128) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 112)];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2156) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 2184)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2184) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 2212)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2212) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 196)];
+ kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2240) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 224)];
+ kernel_shared[(((int)threadIdx.x) + 2268)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2268) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 252)];
+ kernel_shared[(((int)threadIdx.x) + 2296)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2296) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2324)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2324) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 20)];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2352) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 2380)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2380) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 2408)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2408) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 2436)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2436) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2464) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 160)];
+ kernel_shared[(((int)threadIdx.x) + 2492)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2492) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 2520)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2520) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2548) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 244)];
+ kernel_shared[(((int)threadIdx.x) + 2576)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2576) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 272) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2604)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2604) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 12)];
+ kernel_shared[(((int)threadIdx.x) + 2632)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2632) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 40)];
+ kernel_shared[(((int)threadIdx.x) + 2660)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2660) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 68)];
+ kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2688) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 2716)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2716) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 124)];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2744) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 152)];
+ kernel_shared[(((int)threadIdx.x) + 2772)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2772) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 180)];
+ kernel_shared[(((int)threadIdx.x) + 2800)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2800) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 208)];
+ kernel_shared[(((int)threadIdx.x) + 2828)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2828) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 236)];
+ kernel_shared[(((int)threadIdx.x) + 2856)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2856) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 264) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2884)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2884) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 4)];
+ kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2912) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 32)];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2940) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 60)];
+ kernel_shared[(((int)threadIdx.x) + 2968)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2968) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 88)];
+ kernel_shared[(((int)threadIdx.x) + 2996)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2996) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 116)];
+ kernel_shared[(((int)threadIdx.x) + 3024)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3024) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 3052)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3052) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 172)];
+ kernel_shared[(((int)threadIdx.x) + 3080)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3080) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 200)];
+ kernel_shared[(((int)threadIdx.x) + 3108)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3108) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 228)];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3136) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 256)];
+ kernel_shared[(((int)threadIdx.x) + 3164)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3164) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 284) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3192)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3192) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 3220)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3220) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 52)];
+ kernel_shared[(((int)threadIdx.x) + 3248)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3248) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 80)];
+ kernel_shared[(((int)threadIdx.x) + 3276)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3276) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 108)];
+ kernel_shared[(((int)threadIdx.x) + 3304)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3304) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 136)];
+ kernel_shared[(((int)threadIdx.x) + 3332)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3332) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 164)];
+ kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3360) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 192)];
+ kernel_shared[(((int)threadIdx.x) + 3388)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3388) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 220)];
+ kernel_shared[(((int)threadIdx.x) + 3416)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3416) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 248)];
+ kernel_shared[(((int)threadIdx.x) + 3444)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3444) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 276) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3472)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3472) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 16)];
+ kernel_shared[(((int)threadIdx.x) + 3500)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3500) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 44)];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3528) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 3556)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3556) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 100)];
+ kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3584) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 128)];
+ kernel_shared[(((int)threadIdx.x) + 3612)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3612) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 156)];
+ kernel_shared[(((int)threadIdx.x) + 3640)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3640) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 184)];
+ kernel_shared[(((int)threadIdx.x) + 3668)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3668) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 212)];
+ kernel_shared[(((int)threadIdx.x) + 3696)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3696) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 240)];
+ kernel_shared[(((int)threadIdx.x) + 3724)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3724) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 268) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3752)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3752) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 8)];
+ kernel_shared[(((int)threadIdx.x) + 3780)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3780) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 36)];
+ kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3808) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64)];
+ kernel_shared[(((int)threadIdx.x) + 3836)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3836) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 92)];
+ kernel_shared[(((int)threadIdx.x) + 3864)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3864) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 120)];
+ kernel_shared[(((int)threadIdx.x) + 3892)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3892) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 148)];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3920) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 176)];
+ kernel_shared[(((int)threadIdx.x) + 3948)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3948) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 204)];
+ kernel_shared[(((int)threadIdx.x) + 3976)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 3976) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 232)];
+ kernel_shared[(((int)threadIdx.x) + 4004)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4004) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 260)];
+ kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 4060)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4060) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 4088)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4088) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 56)];
+ kernel_shared[(((int)threadIdx.x) + 4116)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4116) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 4144)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4144) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 112)];
+ kernel_shared[(((int)threadIdx.x) + 4172)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4172) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 4200)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4200) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 4228)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4228) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 196)];
+ kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4256) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 224)];
+ kernel_shared[(((int)threadIdx.x) + 4284)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4284) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 252)];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4312) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4340)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4340) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 20)];
+ kernel_shared[(((int)threadIdx.x) + 4368)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4368) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 4396)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4396) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 4424)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4424) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 4452)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4452) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4480) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 160)];
+ kernel_shared[(((int)threadIdx.x) + 4508)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4508) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 4536)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4536) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 4564)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4564) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 244)];
+ if (((int)threadIdx.x) < 16) {
+ kernel_shared[(((int)threadIdx.x) + 4592)] = kernel[(((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 4592) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 272)];
}
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[((((int)threadIdx.x) / 49) * 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 54)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 55)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 56)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 57)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 58)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 59)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 60)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 61)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 62)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 63)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 64)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 65)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 66)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 67)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 68)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 51)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 69)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 52)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 70)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 53)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((((int)threadIdx.x) % 49) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((int)threadIdx.x) / 49) * 72) + 71)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 8; ++rc_outer_inner) {
+ for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 108)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 270)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 117)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 198)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 279)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 207)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 288)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3))]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2304)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 288)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2592)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 9)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2313)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 297)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2601)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 18)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2322)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 306)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2610)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 27)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2331)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 315)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 297)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2619)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 109)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 271)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 37)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 118)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 199)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 280)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 46)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 208)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 289)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 1)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2305)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 289)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 55)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2593)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 10)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2314)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 298)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2602)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 19)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2323)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 307)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2611)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 28)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2332)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 316)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 298)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2620)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 29)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 110)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 272)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 38)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 119)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 200)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 281)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 47)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 209)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 290)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2306)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 290)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 56)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2594)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 11)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2315)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 299)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2603)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 20)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2324)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 308)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2612)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 29)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2333)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 317)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((rc_outer_inner * 324) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 299)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (rc_outer_inner * 36)) + (ry_outer_inner * 3)) + 2621)]));
+ }
+ }
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
- compute[((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 49) * 196)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 49) * 4)) + i1_inner)]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[((((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7)) + 392)] = max((conv2d_nchw[(((i1_inner * 7) + i2_inner) + 14)] + bias[((((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 8)]), 0.000000e+00f);
+ }
}
}
</pre></div>
@@ -884,7 +2279,7 @@ 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 5.440 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 18.394 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 debb5b1dc3..12453f7cf7 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)
- 8.1305 8.1301 8.1391 8.1222 0.0069
+ 8.1169 8.1178 8.1304 8.1025 0.0114
</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 11.032 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 12.362 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 cbd2644833..d4faa035a7 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)
- 751.3321 751.4278 751.6015 750.9669 0.2678
+ 762.6329 762.5277 763.3661 762.0048 0.5607
</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 42.146 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 44.131 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 bec3395a7e..802e663b4f 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -637,74 +637,26 @@ 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(16):
- compute_1 = T.allocate([4096], "float32", "global")
- compute_2 = T.Buffer((4096,), data=compute_1)
- for i_outer_inner, nb_j_inner in T.grid(2, 2):
- for i_inner_init in range(64):
- cse_var_1: T.int32 = i_outer_inner * 2048 + i_inner_init * 32 + nb_j_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, i_inner in T.grid(T.Let(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2], where={cse_var_2: i0_outer_i1_outer_fused * 2 + nb_j_inner}), 64):
- cse_var_2 = T.int32()
+ for i0_outer_i1_outer_fused in T.parallel(32):
+ compute_1 = T.allocate([2048], "float32", "global")
+ compute_2 = T.Buffer((2048,), data=compute_1)
+ for nb_j_inner in range(2):
+ for i_inner_init, j_init in T.grid(64, 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}), 64, 16):
+ cse_var_1 = T.int32()
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_21: T.int32 = elem_idx * 16
- cse_var_20: T.int32 = i0_outer_i1_outer_fused * 2 + nb_j_inner
- cse_var_19: T.int32 = i_outer_inner * 16384 + i_inner * 256
- cse_var_18: T.int32 = i_outer_inner * 2048 + i_inner * 32 + nb_j_inner * 16
- cse_var_17: T.int32 = cse_var_18 + 9
- cse_var_16: T.int32 = cse_var_18 + 8
- cse_var_15: T.int32 = cse_var_18 + 7
- cse_var_14: T.int32 = cse_var_18 + 6
- cse_var_13: T.int32 = cse_var_18 + 5
- cse_var_12: T.int32 = cse_var_18 + 4
- cse_var_11: T.int32 = cse_var_18 + 3
- cse_var_10: T.int32 = cse_var_18 + 2
- cse_var_9: T.int32 = cse_var_18 + 15
- cse_var_8: T.int32 = cse_var_18 + 14
- cse_var_7: T.int32 = cse_var_18 + 13
- cse_var_6: T.int32 = cse_var_18 + 12
- cse_var_5: T.int32 = cse_var_18 + 11
- cse_var_4: T.int32 = cse_var_18 + 10
- cse_var_3: T.int32 = cse_var_18 + 1
+ 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_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 1] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 2] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 3] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 4] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 5] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 6] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 7] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 8] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 9] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 10] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 11] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 12] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 13] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 14] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 15] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
- for i0_inner in range(128):
- cse_var_22: T.int32 = i0_inner * 512 + i0_outer_i1_outer_fused * 32
+ compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 16384 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
+ for i0_inner in range(64):
+ cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 32768 + 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_22:cse_var_22 + 32] = T.max(compute_2[i0_inner * 32:i0_inner * 32 + 32] + placeholder_5[cse_var_22:cse_var_22 + 32], T.Broadcast(T.float32(0), 32))
+ 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))
</pre></div>
</div>
</div>
@@ -738,7 +690,7 @@ 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: 1.862 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.800 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 368d49aa25..afddfdb892 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>00:48.277</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:49.144</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>00:48.243</p></td>
+<td><p>00:49.107</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>00:00.020</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>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>
-<td><p>00:00.004</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>00:00.004</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 991381a696..b5a9be96d4 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -573,8 +573,7 @@ for this template</p>
waiting for device...
device available
Get devices for measurement successfully!
-No: 1 GFLOPS: 84.76/84.76 result: MeasureResult(costs=(0.0027312945238095235,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.309992551803589, timestamp=1680835770.9312594) [('tile_f', [-1, 1, 64, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3717943
-No: 2 GFLOPS: 0.00/84.76 result: Traceback (most recent call last):
+No: 1 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
@@ -696,8 +695,8 @@ 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, 32, 2, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7517548
-No: 3 GFLOPS: 0.00/84.76 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5214228
+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
@@ -819,9 +818,9 @@ 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, 8, 8, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6568130
-No: 4 GFLOPS: 21.67/84.76 result: MeasureResult(costs=(0.010683162272727273,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1216676235198975, timestamp=1680835775.0292473) [('tile_f', [-1, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3292251
-No: 5 GFLOPS: 0.00/84.76 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7951828
+No: 3 GFLOPS: 2.90/2.90 result: MeasureResult(costs=(0.0799303665,), error_no=MeasureErrorNo.NO_ERROR, all_cost=6.104713439941406, timestamp=1680896039.276656) [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4138875
+No: 4 GFLOPS: 0.00/2.90 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
@@ -943,9 +942,8 @@ 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, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2186973
-No: 6 GFLOPS: 135.98/135.98 result: MeasureResult(costs=(0.0017025168611111113,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5132319927215576, timestamp=1680835778.2157695) [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5271275
-No: 7 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3780549
+No: 5 GFLOPS: 0.00/2.90 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
@@ -1067,8 +1065,8 @@ 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, 16, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6674618
-No: 8 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2476650
+No: 6 GFLOPS: 0.00/2.90 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
@@ -1190,9 +1188,8 @@ 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, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9799918
-No: 9 GFLOPS: 1.04/135.98 result: MeasureResult(costs=(0.222601181,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.189843654632568, timestamp=1680835782.5912874) [('tile_f', [-1, 1, 1, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2636680
-No: 10 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 2, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7635467
+No: 7 GFLOPS: 0.00/2.90 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
@@ -1314,8 +1311,8 @@ 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, 64]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9989524
-No: 11 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 128]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3176350
+No: 8 GFLOPS: 0.00/2.90 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
@@ -1437,8 +1434,8 @@ 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, 16, 4, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1547503
-No: 12 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3180035
+No: 9 GFLOPS: 0.00/2.90 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
@@ -1560,8 +1557,8 @@ 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, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9646654
-No: 13 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5995630
+No: 10 GFLOPS: 0.00/2.90 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
@@ -1683,8 +1680,10 @@ 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, 1, 2, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3826664
-No: 14 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6795092
+No: 11 GFLOPS: 208.66/208.66 result: MeasureResult(costs=(0.0011094419000000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.71787428855896, timestamp=1680896046.8706958) [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7009134
+No: 12 GFLOPS: 5.45/208.66 result: MeasureResult(costs=(0.042511791,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.697552680969238, timestamp=1680896047.9853632) [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4612311
+No: 13 GFLOPS: 0.00/208.66 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
@@ -1806,9 +1805,131 @@ 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, 4, 32, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5253292
-No: 15 GFLOPS: 2.56/135.98 result: MeasureResult(costs=(0.0903941345,), error_no=MeasureErrorNo.NO_ERROR, all_cost=6.586085557937622, timestamp=1680835794.574986) [('tile_f', [-1, 8, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4195678
-No: 16 GFLOPS: 0.00/135.98 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, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3692672
+No: 14 GFLOPS: 0.00/208.66 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+ func = build(s, args, target=target, runtime=runtime)
+ File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+ input_mod = lower(inputs, args, name=name, binds=binds)
+ File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+ return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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
+
+Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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, 1, 2, 256]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4278558
+No: 15 GFLOPS: 0.00/208.66 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
res = future.result()
File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1825,9 +1946,131 @@ No: 16 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
raise TimeoutError()
TimeoutError
- [('tile_f', [-1, 128, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8979317
-No: 17 GFLOPS: 15.41/135.98 result: MeasureResult(costs=(0.015019139555555554,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.297766923904419, timestamp=1680835798.044399) [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1429725
-No: 18 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+ [('tile_f', [-1, 128, 1, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,395402
+No: 16 GFLOPS: 0.00/208.66 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+ func = build(s, args, target=target, runtime=runtime)
+ File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+ input_mod = lower(inputs, args, name=name, binds=binds)
+ File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+ return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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
+
+Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc: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, 1, 32, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8635040
+No: 17 GFLOPS: 0.00/208.66 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
@@ -1949,8 +2192,9 @@ 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, 8, 4, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3245022
-No: 19 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1018204
+No: 18 GFLOPS: 49.45/208.66 result: MeasureResult(costs=(0.00468107908,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5602219104766846, timestamp=1680896061.1268966) [('tile_f', [-1, 8, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7199189
+No: 19 GFLOPS: 0.00/208.66 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
@@ -2072,8 +2316,8 @@ 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, 2, 4, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8110476
-No: 20 GFLOPS: 0.00/135.98 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8490621
+No: 20 GFLOPS: 0.00/208.66 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
@@ -2195,7 +2439,7 @@ 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, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559451
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,401453
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2234,9 +2478,9 @@ 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, 8, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5271275
+[('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7009134
Finish loading 20 records
-Time cost of this operator: 0.002048
+Time cost of this operator: 0.001501
</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 08255991cd..1501132ba1 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 312.9 98.715 (1, 2, 10, 10, 3) 2 1 [312.9]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.108 0.98 (1, 6, 10, 10) 1 1 [3.108]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.965 0.304 (1, 1, 10, 10, 3) 1 1 [0.965]
-Total_time - 316.973 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 335.4 98.824 (1, 2, 10, 10, 3) 2 1 [335.4]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.039 0.895 (1, 6, 10, 10) 1 1 [3.039]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.953 0.281 (1, 1, 10, 10, 3) 1 1 [0.953]
+Total_time - 339.392 - - - - -
</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 102.6 97.511 (1, 6, 10, 10, 1) 2 1 [102.6]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.779 1.691 (1, 6, 10, 10) 1 1 [1.779]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.839 0.798 (1, 3, 10, 10, 1) 1 1 [0.839]
-Total_time - 105.219 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.4 97.304 (1, 6, 10, 10, 1) 2 1 [100.4]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.808 1.752 (1, 6, 10, 10) 1 1 [1.808]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.974 0.944 (1, 1, 10, 10, 3) 1 1 [0.974]
+Total_time - 103.182 - - - - -
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.999 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 27.079 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 4148207fae..b1b351c3b6 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -460,7 +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]
-100%|##########| 3.42M/3.42M [00:00<00:00, 191MB/s]
+100%|##########| 3.42M/3.42M [00:00<00:00, 45.8MB/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.
@@ -586,7 +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 16.918 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.594 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 08167a0f7d..0281fdbe76 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/tmp9svv0koq/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmprep4b99f/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="[1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp9svv0koq/images/target contains 8144 images
-/tmp/tmp9svv0koq/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], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [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/tmprep4b99f/images/target contains 8144 images
+/tmp/tmprep4b99f/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -701,13 +701,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 41s - loss: 0.2119 - accuracy: 0.9250 - val_loss: 0.1330 - val_accuracy: 0.9566 - 41s/epoch - 125ms/step
+328/328 - 41s - loss: 0.2114 - accuracy: 0.9269 - val_loss: 0.1611 - val_accuracy: 0.9422 - 41s/epoch - 126ms/step
Epoch 2/3
-328/328 - 34s - loss: 0.1017 - accuracy: 0.9620 - val_loss: 0.1659 - val_accuracy: 0.9358 - 34s/epoch - 104ms/step
+328/328 - 35s - loss: 0.1053 - accuracy: 0.9621 - val_loss: 0.2311 - val_accuracy: 0.9309 - 35s/epoch - 107ms/step
Epoch 3/3
-328/328 - 34s - loss: 0.0594 - accuracy: 0.9768 - val_loss: 0.1022 - val_accuracy: 0.9687 - 34s/epoch - 105ms/step
+328/328 - 35s - loss: 0.0644 - accuracy: 0.9763 - val_loss: 0.1123 - val_accuracy: 0.9619 - 35s/epoch - 107ms/step
-<keras.callbacks.History object at 0x7fb0e80c2550>
+<keras.callbacks.History object at 0x7f2b973ad150>
</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 16.347 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 25.309 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 a9a178febd..0a2eb99144 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>07:22.699</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>07:39.656</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>04:16.347</p></td>
+<td><p>04:25.309</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">6. Model Tuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>01:23.999</p></td>
+<td><p>01:27.079</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>01:16.918</p></td>
+<td><p>01:20.594</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>00:10.271</p></td>
+<td><p>00:10.618</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>00:07.780</p></td>
+<td><p>00:08.383</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>00:07.385</p></td>
+<td><p>00:07.674</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 f2d50e1326..c64d097526 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>00:36.659</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:37.759</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>00:32.015</p></td>
+<td><p>00:32.981</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></td>
-<td><p>00:02.860</p></td>
+<td><p>00:02.971</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>00:01.778</p></td>
+<td><p>00:01.800</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 c1f83e7570..2013e418ed 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 0x7fb111e3ba70>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f27d29a0a70>
</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 420518defa..9f1adef6b9 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>00:07.031</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:08.666</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>00:04.205</p></td>
+<td><p>00:05.849</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>00:01.317</p></td>
+<td><p>00:01.287</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>00:00.621</p></td>
+<td><p>00:00.628</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>00:00.614</p></td>
+<td><p>00:00.616</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>00:00.127</p></td>
+<td><p>00:00.133</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>00:00.062</p></td>
+<td><p>00:00.065</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.055</p></td>
+<td><p>00:00.056</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></td>
-<td><p>00:00.030</p></td>
+<td><p>00:00.031</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index e285b427a5..1515c4b747 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -234,17 +234,7 @@
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
-<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
-<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
-</ul>
-</li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
-</ul>
-</li>
+<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
<li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index c39a497dde..5031572192 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1621,7 +1621,7 @@ history states as starting point to perform Evolutionary Search).</p></li>
<dl class="py class">
<dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
<dd><p>The search policy that searches in a hierarchical search space defined by sketches.
The policy randomly samples programs from the space defined by sketches and use evolutionary
search to fine-tune them.</p>
@@ -1905,7 +1905,7 @@ Candidates:
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index b7c34cc976..b64457a45a 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/b228037a2/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 96b6a7f09d..a69cebe715 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/b228037a2/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 bdf65e54e1..a36368ab49 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/b228037a2/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L357">runtime.ts:357</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L376">runtime.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L367">runtime.ts:367</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 56d0a4c0c5..1dc0f2e544 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/b228037a2/web/src/runtime.ts#L299">runtime.ts:299</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L320">runtime.ts:320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L327">runtime.ts:327</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 203a1a4bf9..42e5dd90f4 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/b228037a2/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 5c106efa84..8237cf7589 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/b228037a2/web/src/runtime.ts#L50">runtime.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L48">runtime.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L77">runtime.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L67">runtime.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L85">runtime.ts:85</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L96">runtime.ts:96</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L73">runtime.ts:73</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 cc89a7a7a3..bef2b7c074 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/b228037a2/web/src/runtime.ts#L844">runtime.ts:844</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L834">runtime.ts:834</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L833">runtime.ts:833</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L973">runtime.ts:973</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L901">runtime.ts:901</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L922">runtime.ts:922</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L943">runtime.ts:943</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L986">runtime.ts:986</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L957">runtime.ts:957</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 d84b2527ea..c9900748ca 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/b228037a2/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 20542fb5ba..6f5681d94e 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/b228037a2/web/src/runtime.ts#L614">runtime.ts:614</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L626">runtime.ts:626</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L653">runtime.ts:653</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L641">runtime.ts:641</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L687">runtime.ts:687</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 644b2f3e0c..85048c8a71 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/b228037a2/web/src/runtime.ts#L401">runtime.ts:401</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L394">runtime.ts:394</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L390">runtime.ts:390</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L388">runtime.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L392">runtime.ts:392</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L480">runtime.ts:480</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L524">runtime.ts:524</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L465">runtime.ts:465</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L458">runtime.ts:458</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L584">runtime.ts:584</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L553">runtime.ts:553</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 4eeb5b9c4b..65fcdf4d48 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/b228037a2/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L255">runtime.ts:255</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L264">runtime.ts:264</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 2acba4a5ce..9fc98d2eb4 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/b228037a2/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 bb873c48aa..477c5123a5 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/b228037a2/web/src/runtime.ts#L148">runtime.ts:148</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L144">runtime.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L146">runtime.ts:146</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L163">runtime.ts:163</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L208">runtime.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L167">runtime.ts:167</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 cb7b9d9095..26fbf3ae2d 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/b228037a2/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L233">runtime.ts:233</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 b5e56cb7ba..35c2a089f1 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/b228037a2/web/src/runtime.ts#L784">runtime.ts:784</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L804">runtime.ts:804</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L796">runtime.ts:796</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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/b228037a2/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/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 703353a00d..756e823d96 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/b228037a2/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">ctx<span class="tsd-signature-symbol">:</span> <a href="runtimecontext.html" class="tsd-signature-type">RuntimeContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -175,7 +175,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -192,7 +192,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -246,7 +246,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/webgpucontext.html b/docs/reference/api/typedoc/classes/webgpucontext.html
index 73a9de7a02..43c278b312 100644
--- a/docs/reference/api/typedoc/classes/webgpucontext.html
+++ b/docs/reference/api/typedoc/classes/webgpucontext.html
@@ -120,7 +120,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -145,7 +145,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">GPUDevice</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
</ul>
</aside>
</section>
@@ -155,7 +155,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
</ul>
</aside>
</section>
@@ -172,7 +172,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -209,7 +209,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/enums/argtypecode.html b/docs/reference/api/typedoc/enums/argtypecode.html
index 642ec7a86d..c13e57152c 100644
--- a/docs/reference/api/typedoc/enums/argtypecode.html
+++ b/docs/reference/api/typedoc/enums/argtypecode.html
@@ -106,7 +106,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 6</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
</ul>
</aside>
</section>
@@ -116,7 +116,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
</ul>
</aside>
</section>
@@ -126,7 +126,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
</ul>
</aside>
</section>
@@ -136,7 +136,7 @@
<div class="tsd-signature tsd-kind-icon">Null<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
</ul>
</aside>
</section>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">TVMBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 12</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
</ul>
</aside>
</section>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">TVMDLTensor<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">TVMData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">TVMModule<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 9</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">TVMNDArray<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 13</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
</ul>
</aside>
</section>
@@ -196,7 +196,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
</ul>
</aside>
</section>
@@ -206,7 +206,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObjectRValue<wbr>Ref<wbr>Arg<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 14</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
</ul>
</aside>
</section>
@@ -216,7 +216,7 @@
<div class="tsd-signature tsd-kind-icon">TVMOpaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
</ul>
</aside>
</section>
@@ -226,7 +226,7 @@
<div class="tsd-signature tsd-kind-icon">TVMPacked<wbr>Func<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 10</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
</ul>
</aside>
</section>
@@ -236,7 +236,7 @@
<div class="tsd-signature tsd-kind-icon">TVMStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 11</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
</ul>
</aside>
</section>
@@ -246,7 +246,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index e0279ae41c..0fb0cb04cd 100644
--- a/docs/reference/api/typedoc/enums/aynccallbackcode.html
+++ b/docs/reference/api/typedoc/enums/aynccallbackcode.html
@@ -93,7 +93,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Exception<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L812">runtime.ts:812</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L812">runtime.ts:812</a></li>
</ul>
</aside>
</section>
@@ -103,7 +103,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L811">runtime.ts:811</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L811">runtime.ts:811</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index 03a74d8d47..d99345c716 100644
--- a/docs/reference/api/typedoc/enums/dldatatypecode.html
+++ b/docs/reference/api/typedoc/enums/dldatatypecode.html
@@ -95,7 +95,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L339">runtime.ts:339</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L339">runtime.ts:339</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L337">runtime.ts:337</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L337">runtime.ts:337</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">Opaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L340">runtime.ts:340</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L340">runtime.ts:340</a></li>
</ul>
</aside>
</section>
@@ -125,7 +125,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L338">runtime.ts:338</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L338">runtime.ts:338</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index 6c4d51c639..2d018c706b 100644
--- a/docs/reference/api/typedoc/enums/rpcserverstate.html
+++ b/docs/reference/api/typedoc/enums/rpcserverstate.html
@@ -90,7 +90,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
</ul>
</aside>
</section>
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<wbr>Key<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Server<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Body<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">Wait<wbr>For<wbr>Callback<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index 1950ce3357..2e252fa0b6 100644
--- a/docs/reference/api/typedoc/enums/sizeof.html
+++ b/docs/reference/api/typedoc/enums/sizeof.html
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32 + I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">F32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">F64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">I32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
</ul>
</aside>
</section>
@@ -150,7 +150,7 @@
<div class="tsd-signature tsd-kind-icon">I64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
</ul>
</aside>
</section>
@@ -160,7 +160,7 @@
<div class="tsd-signature tsd-kind-icon">TVMValue<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
</ul>
</aside>
</section>
@@ -170,7 +170,7 @@
<div class="tsd-signature tsd-kind-icon">U16<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<div class="tsd-signature tsd-kind-icon">U8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 159c57442c..f1e47f9b0d 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">FObject<wbr>Constructor<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, lib<span class="tsd-signature-symbol">: </span><a href="classes/ffilibrary.html" class="tsd-signature-type">FFILibrary</a>, ctx<span class="tsd-signature-symbol">: </span><a href="classes/runtimecontext.html" class="t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L778">runtime.ts:778</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L778">runtime.ts:778</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Alloc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>shape<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, ndim<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeCode<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeBits<span class="tsd [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -288,7 +288,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>Bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">num [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>To<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>from<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, to<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-sig [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -376,7 +376,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>ToBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</sp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -420,7 +420,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/b228037a2/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -456,7 +456,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMBackend<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number< [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -508,7 +508,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCFunc<wbr>Set<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ret<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -556,7 +556,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCb<wbr>Arg<wbr>ToReturn<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, code<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span c [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -595,7 +595,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Call<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -651,7 +651,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/b228037a2/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -687,7 +687,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Get<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span cla [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -726,7 +726,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>List<wbr>Global<wbr>Names<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>outSize<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, outArray<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -765,7 +765,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Register<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, f<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, override<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -808,7 +808,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMGet<wbr>Last<wbr>Error<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -838,7 +838,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/b228037a2/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -874,7 +874,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Get<wbr>Function<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, funcName<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, queryImports<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">numbe [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -922,7 +922,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Import<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, dep<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-si [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -962,7 +962,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/b228037a2/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -998,7 +998,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Get<wbr>Type<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1037,7 +1037,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Index2<wbr>Key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_index<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, out_type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1076,7 +1076,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Key2<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1115,7 +1115,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMSynchronize<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>deviceType<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, deviceId<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signatur [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1157,7 +1157,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Alloc<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>size<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1193,7 +1193,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Free<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ptr<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/b228037a2/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1229,7 +1229,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Func<wbr>Create<wbr>FromCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resource<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1269,7 +1269,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>args<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1321,7 +1321,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<wbr>Finalizer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resourceHandle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </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/b228037a2/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1357,7 +1357,7 @@
<div class="tsd-signature tsd-kind-icon">GPUPointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1372,7 +1372,7 @@
<div class="tsd-signature tsd-kind-icon">Packed<wbr>Func<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">...</span>args<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol"> & </span><a href="interfaces/disp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L37">runtime.ts:37</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L37">runtime.ts:37</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1387,7 +1387,7 @@
<div class="tsd-signature tsd-kind-icon">Pointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1402,7 +1402,7 @@
<div class="tsd-signature tsd-kind-icon">Ptr<wbr>Offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1417,7 +1417,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Base<span class="tsd-signature-symbol">:</span> <a href="classes/tvmobject.html" class="tsd-signature-type">TVMObject</a><span class="tsd-signature-symbol"> | </span><a href="classes/ndarray.html" class="tsd-signature-type">NDArray</a><span class="tsd-signature-symbol"> | </span><a href="classes/module.html" class="tsd-signature-type">Module</a><span class="tsd-signature-symbol"> | </span><a href="index.html#packedfunc" class="t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/runtime.ts#L781">runtime.ts:781</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/runtime.ts#L781">runtime.ts:781</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1435,7 +1435,7 @@
<div class="tsd-signature tsd-kind-icon">RPC_<wbr>MAGIC<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">1045105</span><span class="tsd-signature-symbol"> = 1045105</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1457,7 +1457,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/support.ts#L25">support.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/support.ts#L25">support.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1489,7 +1489,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/b228037a2/web/src/support.ts#L39">support.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/460374fed/web/src/support.ts#L39">support.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1518,7 +1518,7 @@
<li class="tsd-description">
... 1083 lines suppressed ...