[LLM] First commit the llm deployment code

custom_ops/gpu_ops/save_with_output_msg.h

@@ -0,0 +1,39 @@
+// 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.
+
+#pragma once
+
+#include <stdio.h>
+#include <string.h>
+#include <sys/ipc.h>
+#include <sys/msg.h>
+#include <sys/types.h>
+#include "paddle/extension.h"
+
+#ifndef PD_BUILD_STATIC_OP
+#define PD_BUILD_STATIC_OP(name) PD_BUILD_OP(static_op_##name)
+#endif
+
+#define MAX_BSZ 512
+
+struct msgdata {
+    long mtype;
+    int mtext[MAX_BSZ + 2];  // stop_flag, bsz, tokens
+};
+
+void SaveOutMmsg(const paddle::Tensor& x,
+                 const paddle::Tensor& not_need_stop,
+                 int64_t rank_id,
+                 int msg_queue_id,
+                 bool save_each_rank);