xref: /aosp_15_r20/external/tensorflow/tensorflow/core/common_runtime/gpu/gpu_cudamalloc_allocator.h (revision b6fb3261f9314811a0f4371741dbb8839866f948)
1 /* Copyright 2017 The TensorFlow Authors. All Rights Reserved.
2 
3 Licensed under the Apache License, Version 2.0 (the "License");
4 you may not use this file except in compliance with the License.
5 You may obtain a copy of the License at
6 
7     http://www.apache.org/licenses/LICENSE-2.0
8 
9 Unless required by applicable law or agreed to in writing, software
10 distributed under the License is distributed on an "AS IS" BASIS,
11 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12 See the License for the specific language governing permissions and
13 limitations under the License.
14 ==============================================================================*/
15 
16 #ifndef TENSORFLOW_CORE_COMMON_RUNTIME_GPU_GPU_CUDAMALLOC_ALLOCATOR_H_
17 #define TENSORFLOW_CORE_COMMON_RUNTIME_GPU_GPU_CUDAMALLOC_ALLOCATOR_H_
18 
19 #include <memory>
20 
21 #include "tensorflow/core/common_runtime/gpu/gpu_id.h"
22 #include "tensorflow/core/framework/allocator.h"
23 #include "tensorflow/core/platform/macros.h"
24 #include "tensorflow/core/platform/stream_executor.h"
25 #include "tensorflow/core/platform/types.h"
26 
27 namespace tensorflow {
28 
29 // An allocator which directly uses cuMemAlloc and cuMemFree to allocate and
30 // free memory.
31 class GPUcudaMallocAllocator : public Allocator {
32  public:
33   explicit GPUcudaMallocAllocator(PlatformDeviceId platform_device_id);
Name()34   string Name() override { return "gpu_debug"; }
35   void* AllocateRaw(size_t alignment, size_t num_bytes) override;
36   void DeallocateRaw(void* ptr) override;
37   bool TracksAllocationSizes() const override;
38 
GetMemoryType()39   AllocatorMemoryType GetMemoryType() const override {
40     return AllocatorMemoryType::kDevice;
41   }
42 
43  private:
44   se::StreamExecutor* stream_exec_;  // Not owned.
45 
46   TF_DISALLOW_COPY_AND_ASSIGN(GPUcudaMallocAllocator);
47 };
48 
49 }  // namespace tensorflow
50 
51 #endif  // TENSORFLOW_CORE_COMMON_RUNTIME_GPU_GPU_CUDAMALLOC_ALLOCATOR_H_
52