[Inference] Split core converter file into several files (#67877)

* split core

* fix bug

* delete comment

* delete code

Author: YuanRisheng

paddle/fluid/inference/tensorrt/pir/generic_plugin.cu

@@ -573,6 +573,10 @@ int GenericPlugin::enqueue(const nvinfer1::PluginTensorDesc* input_desc,
         (data_type == nvinfer1::DataType::kHALF));
 
   phi_kernel_contexts_[data_type]->ClearInputOutput();
+
+  auto* dev_ctx = static_cast<phi::GPUContext*>(pool.Get(place));
+  phi_kernel_contexts_[data_type]->SetDeviceContext(dev_ctx);
+
   auto& vec_kernel_fn_tensor_params = op_yaml_info_->TensorParams(true);
   int kernel_input_count = vec_kernel_fn_tensor_params.size();
   for (int i = 0; i < getNbInputs(); i++) {

python/paddle/tensorrt/converter.py

@@ -23,7 +23,15 @@
 from paddle.base.core import get_value_shape_range_info
 from paddle.base.log_helper import get_logger
 
-from .impls.core import *  # noqa: F403
+from .impls.activation import *  # noqa: F403
+from .impls.conv import *  # noqa: F403
+from .impls.creation import *  # noqa: F403
+from .impls.linalg import *  # noqa: F403
+from .impls.manipulation import *  # noqa: F403
+from .impls.math import *  # noqa: F403
+from .impls.norm import *  # noqa: F403
+from .impls.pooling import *  # noqa: F403
+from .impls.search import *  # noqa: F403
 from .register import converter_registry
 from .util import map_dtype
 
@@ -200,12 +208,11 @@ def convert_subgraph_to_trt(self, program, group_op):
                             f'{source_id} not found in value_to_trt_tensor'
                         )
 
-            layer = self.convert(network, op, operands)
+            trt_outs = self.convert(network, op, operands)
 
             for idx, result in enumerate(op.results()):
-                # TODO In some cases, the output index (idx) of a Paddle OP may not necessarily be the same as the output index of TensorRT
-                if idx < layer.num_outputs:
-                    value_to_trt_tensor[result.id] = layer.get_output(idx)
+                if idx < len(trt_outs):
+                    value_to_trt_tensor[result.id] = trt_outs[idx]
                 else:
                     value_to_trt_tensor[result.id] = None
         out_shapes = []
@@ -307,8 +314,11 @@ def convert(self, network, paddle_op, inputs):
                 raise NotImplementedError(
                     f"Converter for {op_name} not implemented."
                 )
-            out = converter_func(network, paddle_op, inputs)
-        return out
+            outs = converter_func(network, paddle_op, inputs)
+        if isinstance(outs, tuple):
+            return outs
+        else:
+            return tuple(outs)
 
     def convert_program_to_trt(self):
         for op in self.program.global_block().ops:

python/paddle/tensorrt/impls/activation.py

@@ -0,0 +1,75 @@
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import logging
+import os
+import sys
+
+import numpy as np
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+parent_dir = os.path.abspath(os.path.join(current_dir, os.pardir))
+if parent_dir not in sys.path:
+    sys.path.append(parent_dir)
+
+import tensorrt as trt
+
+from paddle.base.log_helper import get_logger
+from paddle.tensorrt.register import converter_registry
+
+_logger = get_logger(
+    __name__, logging.INFO, fmt='%(asctime)s-%(levelname)s: %(message)s'
+)
+from paddle.tensorrt.converter_utils import (
+    get_trt_plugin,
+)
+
+
+@converter_registry.register("pd_op.relu", trt_version="8.x")
+def relu_converter(network, paddle_op, inputs):
+    relu_layer = network.add_activation(inputs[0], trt.ActivationType.RELU)
+    return relu_layer.get_output(0)
+
+
+@converter_registry.register("pd_op.softmax", trt_version="8.x")
+def softmax_converter(network, paddle_op, inputs):
+    axis = paddle_op.attrs().get("axis", 0)
+    if axis < 0:
+        axis = len(inputs[0].shape) + axis
+
+    softmax_layer = network.add_softmax(inputs[0])
+    softmax_layer.axes = 1 << axis
+    return softmax_layer.get_output(0)
+
+
+@converter_registry.register("pd_op.gelu", trt_version="8.x")
+def gelu_converter(network, paddle_op, inputs):
+    input_val = inputs[0]
+    approximate = paddle_op.attrs()["approximate"]
+    if approximate:
+        raise RuntimeError(
+            "GeLU converter currently doesn't support fast gelu compute"
+        )
+
+    plugin_name = "CustomGeluPluginDynamic"
+    type_id = trt.PluginField(
+        "type_id", np.array(0, dtype=np.int32), trt.PluginFieldType.INT32
+    )
+
+    filed_collection = trt.PluginFieldCollection([type_id])
+    plugin_version = "1"
+
+    plugin = get_trt_plugin(plugin_name, filed_collection, plugin_version)
+
+    layer = network.add_plugin_v2([input_val], plugin)
+    return layer.get_output(0)

python/paddle/tensorrt/impls/conv.py

@@ -0,0 +1,54 @@
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import logging
+import os
+import sys
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+parent_dir = os.path.abspath(os.path.join(current_dir, os.pardir))
+if parent_dir not in sys.path:
+    sys.path.append(parent_dir)
+
+import tensorrt as trt
+
+from paddle.base.log_helper import get_logger
+from paddle.tensorrt.register import converter_registry
+
+_logger = get_logger(
+    __name__, logging.INFO, fmt='%(asctime)s-%(levelname)s: %(message)s'
+)
+
+
+@converter_registry.register("pd_op.conv2d", trt_version="8.x")
+def conv2d_converter(network, paddle_op, inputs):
+    input_tensor, weight = inputs
+    weight_shape = paddle_op.operands()[1].source().shape
+
+    padding = paddle_op.attrs().get("paddings", [0, 0])
+    stride = paddle_op.attrs().get("strides", [1, 1])
+    dilation = paddle_op.attrs().get("dilations", [1, 1])
+    groups = paddle_op.attrs().get("groups", 1)
+
+    # weight_tensor = network.add_constant(weight_shape, weight).get_output(0)
+    kernel_shape = trt.Dims((weight_shape[2], weight_shape[3]))
+
+    conv_layer = network.add_convolution_nd(
+        input_tensor, weight_shape[0], kernel_shape, weight
+    )
+    conv_layer.stride_nd = stride
+    conv_layer.padding_nd = padding
+    conv_layer.dilation_nd = dilation
+    conv_layer.num_groups = groups
+
+    return conv_layer.get_output(0)