Home
last modified time | relevance | path

Searched refs:TORCH_CUDA_CPP_API (Results 1 – 24 of 24) sorted by relevance

/aosp_15_r20/external/pytorch/torch/csrc/cuda/
H A Dnccl.h90 TORCH_CUDA_CPP_API void throw_nccl_error(ncclResult status);
98 TORCH_CUDA_CPP_API at::ArrayRef<ncclComm_t> get_communicators(
100 TORCH_CUDA_CPP_API void check_inputs(
105 TORCH_CUDA_CPP_API void check_inputs(
117 TORCH_CUDA_CPP_API std::uint64_t version();
118 TORCH_CUDA_CPP_API const char* version_suffix();
122 TORCH_CUDA_CPP_API void get_unique_id(ncclUniqueId& id);
123 TORCH_CUDA_CPP_API ncclComm_t
125 TORCH_CUDA_CPP_API void comm_destroy(ncclComm_t comm);
127 TORCH_CUDA_CPP_API void broadcast(
[all …]
H A DCUDAPluggableAllocator.h22 struct TORCH_CUDA_CPP_API CUDAPluggableAllocatorDeleterContext {
46 TORCH_CUDA_CPP_API std::shared_ptr<
49 TORCH_CUDA_CPP_API std::shared_ptr<
54 TORCH_CUDA_CPP_API void changeCurrentAllocator(
69 struct TORCH_CUDA_CPP_API CUDAPluggableAllocator
/aosp_15_r20/external/pytorch/aten/src/ATen/cuda/
H A DCUDAContextLight.h72 TORCH_CUDA_CPP_API cudaDeviceProp* getCurrentDeviceProperties();
74 TORCH_CUDA_CPP_API int warp_size();
76 TORCH_CUDA_CPP_API cudaDeviceProp* getDeviceProperties(c10::DeviceIndex device);
78 TORCH_CUDA_CPP_API bool canDeviceAccessPeer(
82 TORCH_CUDA_CPP_API c10::Allocator* getCUDADeviceAllocator();
85 TORCH_CUDA_CPP_API cusparseHandle_t getCurrentCUDASparseHandle();
86 TORCH_CUDA_CPP_API cublasHandle_t getCurrentCUDABlasHandle();
87 TORCH_CUDA_CPP_API cublasLtHandle_t getCurrentCUDABlasLtHandle();
89 TORCH_CUDA_CPP_API void clearCublasWorkspaces();
92 TORCH_CUDA_CPP_API cusolverDnHandle_t getCurrentCUDASolverDnHandle();
[all …]
H A DCUDASparseDescriptors.h80 class TORCH_CUDA_CPP_API CuSparseMatDescriptor
104 class TORCH_CUDA_CPP_API CuSparseBsrsv2Info
114 class TORCH_CUDA_CPP_API CuSparseBsrsm2Info
131 class TORCH_CUDA_CPP_API CuSparseDnMatDescriptor
137 class TORCH_CUDA_CPP_API CuSparseConstDnMatDescriptor
149 class TORCH_CUDA_CPP_API CuSparseDnVecDescriptor
155 class TORCH_CUDA_CPP_API CuSparseSpMatDescriptor
159 class TORCH_CUDA_CPP_API CuSparseDnMatDescriptor
169 class TORCH_CUDA_CPP_API CuSparseConstDnMatDescriptor
185 class TORCH_CUDA_CPP_API CuSparseDnVecDescriptor
[all …]
H A DEmptyTensor.h6 TORCH_CUDA_CPP_API TensorBase empty_cuda(
12 TORCH_CUDA_CPP_API TensorBase empty_cuda(
20 TORCH_CUDA_CPP_API TensorBase empty_cuda(
24 TORCH_CUDA_CPP_API TensorBase empty_strided_cuda(
30 TORCH_CUDA_CPP_API TensorBase empty_strided_cuda(
38 TORCH_CUDA_CPP_API TensorBase empty_strided_cuda(
H A Dllvm_jit_strings.h8 TORCH_CUDA_CPP_API const std::string &get_traits_string();
9 TORCH_CUDA_CPP_API const std::string &get_cmath_string();
10 TORCH_CUDA_CPP_API const std::string &get_complex_body_string();
11 TORCH_CUDA_CPP_API const std::string &get_complex_half_body_string();
12 TORCH_CUDA_CPP_API const std::string &get_complex_math_string();
H A DCachingHostAllocator.h21 TORCH_CUDA_CPP_API c10::Allocator* getCachingHostAllocator();
25 TORCH_CUDA_CPP_API bool CachingHostAllocator_recordEvent(
31 TORCH_CUDA_CPP_API void CachingHostAllocator_emptyCache();
33 inline TORCH_CUDA_CPP_API at::DataPtr HostAlloc(size_t size) { in HostAlloc()
H A DCUDAGeneratorImpl.h126 struct TORCH_CUDA_CPP_API CUDAGeneratorImpl : public c10::GeneratorImpl {
176 TORCH_CUDA_CPP_API const Generator& getDefaultCUDAGenerator(
178 TORCH_CUDA_CPP_API Generator createCUDAGenerator(DeviceIndex device_index = -1);
H A Djiterator.h15 TORCH_CUDA_CPP_API c10::SmallVector<at::Tensor> CompileAndLaunchKernel(
29 TORCH_CUDA_CPP_API c10::SmallVector<at::Tensor> CompileAndLaunchKernel( in CompileAndLaunchKernel()
H A DCUDAGraph.h19 TORCH_CUDA_CPP_API MempoolId_t graph_pool_handle();
21 struct TORCH_CUDA_CPP_API CUDAGraph {
H A DPinnedMemoryAllocator.h8 inline TORCH_CUDA_CPP_API at::Allocator* getPinnedMemoryAllocator() { in getPinnedMemoryAllocator()
H A DPeerToPeerAccess.h9 TORCH_CUDA_CPP_API bool get_p2p_access(int source_dev, int dest_dev);
H A DCUDAEvent.h28 struct TORCH_CUDA_CPP_API CUDAEvent {
/aosp_15_r20/external/pytorch/aten/src/ATen/cudnn/
H A DDescriptors.h95 class TORCH_CUDA_CPP_API Descriptor {
123 class TORCH_CUDA_CPP_API RNNDataDescriptor : public Descriptor<
135 class TORCH_CUDA_CPP_API TensorDescriptor : public Descriptor<
176 class TORCH_CUDA_CPP_API FilterDescriptor : public Descriptor<
196 struct TORCH_CUDA_CPP_API ConvolutionDescriptor
217 struct TORCH_CUDA_CPP_API SpatialTransformerDescriptor
227 struct TORCH_CUDA_CPP_API DropoutDescriptor
266 struct TORCH_CUDA_CPP_API RNNDescriptor : public Descriptor<
343 struct TORCH_CUDA_CPP_API CTCLossDescriptor
378 struct TORCH_CUDA_CPP_API ActivationDescriptor
H A DHandle.h8 TORCH_CUDA_CPP_API cudnnHandle_t getCudnnHandle();
H A DTypes.h8 TORCH_CUDA_CPP_API cudnnDataType_t
/aosp_15_r20/external/pytorch/aten/src/ATen/cuda/tunable/
H A DTunable.h64 enum TORCH_CUDA_CPP_API TuningStatus {
71 class TORCH_CUDA_CPP_API ResultEntry {
91 struct TORCH_CUDA_CPP_API TuningResults {
99 class TORCH_CUDA_CPP_API TuningResultsManager {
137 class TORCH_CUDA_CPP_API TuningResultsValidator {
161 class TORCH_CUDA_CPP_API TuningContext {
232 TORCH_CUDA_CPP_API TuningContext* getTuningContext();
/aosp_15_r20/external/pytorch/c10/macros/
H A DExport.h126 #define TORCH_CUDA_CPP_API C10_EXPORT macro
128 #define TORCH_CUDA_CPP_API C10_IMPORT macro
134 #define TORCH_CUDA_CPP_API C10_EXPORT macro
137 #define TORCH_CUDA_CPP_API C10_IMPORT macro
/aosp_15_r20/external/pytorch/aten/src/ATen/native/cudnn/
H A DRNNUtils.h13 TORCH_CUDA_CPP_API std::tuple<Tensor, std::vector<Tensor>>
H A DRNN.cpp1238 TORCH_CUDA_CPP_API std::tuple<Tensor, std::vector<Tensor>>
/aosp_15_r20/external/pytorch/aten/src/ATen/native/cuda/
H A DResize.h10 TORCH_CUDA_CPP_API void resize_bytes_cuda(StorageImpl* storage, size_t size_bytes);
H A Djit_utils.h213 TORCH_CUDA_CPP_API void initializeCudaContext();
/aosp_15_r20/external/pytorch/aten/src/ATen/cuda/detail/
H A DCUDAHooks.h16 TORCH_CUDA_CPP_API void set_magma_init_fn(void (*magma_init_fn)());
/aosp_15_r20/external/pytorch/aten/src/ATen/cuda/nvrtc_stub/
H A DATenNVRTC.h134 extern "C" TORCH_CUDA_CPP_API NVRTC* load_nvrtc();