| /external/tensorflow/tensorflow/stream_executor/ |
| D | device_description.h | 37 struct CudaComputeCapability { struct
44 CudaComputeCapability() {} in CudaComputeCapability() argument
45 CudaComputeCapability(int major, int minor) { in CudaComputeCapability() function
51 return !(*this < CudaComputeCapability{other_major, other_minor});
54 bool operator<(const CudaComputeCapability &other) const {
58 bool operator==(const CudaComputeCapability &other) const {
62 bool operator!=(const CudaComputeCapability &other) const {
171 CudaComputeCapability cuda_compute_capability() const;
244 CudaComputeCapability cuda_compute_capability_{-1, -1};
336 CudaComputeCapability{major, minor}; in set_cuda_compute_capability()
|
| D | device_description.cc | 111 CudaComputeCapability DeviceDescription::cuda_compute_capability() const { in cuda_compute_capability()
|
| D | dnn.cc | 52 CudaComputeCapability cuda_compute_capability, in GetConvolveAlgorithms()
86 CudaComputeCapability cuda_compute_capability, in GetConvolveBackwardDataAlgorithms()
92 CudaComputeCapability cuda_compute_capability, in GetConvolveBackwardFilterAlgorithms()
|
| /external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
| D | ir_emitter_context.h | 42 se::CudaComputeCapability cuda_compute_capability, in IrEmitterContext()
64 se::CudaComputeCapability cuda_compute_capability() const { in cuda_compute_capability()
92 se::CudaComputeCapability cuda_compute_capability_;
|
| D | cudnn_vectorize_convolutions.h | 49 se::CudaComputeCapability compute_capability) in CudnnVectorizeConvolutions()
58 se::CudaComputeCapability compute_capability_;
|
| D | cudnn_pad_for_convolutions.h | 34 explicit CudnnPadForConvolutions(se::CudaComputeCapability compute_capability) in CudnnPadForConvolutions()
44 const se::CudaComputeCapability compute_capability_;
|
| D | tree_reduction_rewriter.h | 79 se::CudaComputeCapability cuda_compute_capability) in GpuTreeReductionRewriter()
90 se::CudaComputeCapability cuda_compute_capability_;
|
| D | nvptx_compiler.cc | 168 se::CudaComputeCapability::AMPERE)) { in OptimizeHloPostLayoutAssignment()
173 se::CudaComputeCapability::VOLTA)) { in OptimizeHloPostLayoutAssignment()
393 stream_exec, ptx, absl::get<se::CudaComputeCapability>(gpu_version), in CompileTargetBinary()
402 se::CudaComputeCapability cc, const HloModuleConfig& hlo_module_config, in CompileGpuAsmOrGetCachedResult()
|
| D | gpu_types.h | 37 using GpuVersion = absl::variant<se::CudaComputeCapability, std::string>;
|
| D | tree_reduction_rewriter.cc | 49 se::CudaComputeCapability cuda_compute_capability) in ReductionRewriterVisitor()
220 se::CudaComputeCapability cuda_compute_capability_;
|
| D | gpu_executable.cc | 135 TF_RET_CHECK(absl::get<se::CudaComputeCapability>(cc) == in CheckCompatibilityWithServiceExecutableRunOptions()
136 absl::get<se::CudaComputeCapability>(gpu_version_)) in CheckCompatibilityWithServiceExecutableRunOptions()
138 << absl::get<se::CudaComputeCapability>(gpu_version_).ToString() in CheckCompatibilityWithServiceExecutableRunOptions()
139 << "}, but was {" << absl::get<se::CudaComputeCapability>(cc).ToString() in CheckCompatibilityWithServiceExecutableRunOptions()
|
| D | nvptx_compiler.h | 82 se::CudaComputeCapability cc, const HloModuleConfig& hlo_module_config,
|
| D | ir_emission_utils.h | 235 se::CudaComputeCapability cuda_compute_capability);
|
| D | gpu_compiler.h | 152 se::CudaComputeCapability cuda_compute_capability, int pointer_size);
|
| D | gpu_compiler.cc | 218 .IsAtLeast(se::CudaComputeCapability::AMPERE); in IsConvBF16Supported()
253 .IsAtLeast(se::CudaComputeCapability::VOLTA) || in OptimizeHloModule()
745 se::CudaComputeCapability cuda_compute_capability, in CompileModuleToLlvmIrImpl()
1177 se::CudaComputeCapability cuda_compute_capability, int pointer_size) { in CompileModuleToLlvmIr()
|
| /external/tensorflow/tensorflow/core/profiler/protobuf/ |
| D | hardware_types.proto | 19 message CudaComputeCapability { message
29 CudaComputeCapability compute_capability = 5;
|
| /external/tensorflow/tensorflow/compiler/xla/service/gpu/tests/ |
| D | reduction_vectorization_test.cc | 85 .IsAtLeast(se::CudaComputeCapability::VOLTA)) { in TEST_F()
|
| D | gemm_rewrite_test.cc | 55 se::CudaComputeCapability GetCudaComputeCapability() { in GetCudaComputeCapability()
536 if (GetCudaComputeCapability().IsAtLeast(se::CudaComputeCapability::VOLTA)) {
564 if (GetCudaComputeCapability().IsAtLeast(se::CudaComputeCapability::VOLTA)) {
|
| D | hlo_to_llvm_ir.cc | 66 tensorflow::se::CudaComputeCapability cuda_compute_capability; in CompileAndPrintLlvmIr()
|
| /external/tensorflow/tensorflow/core/common_runtime/gpu/ |
| D | gpu_device_test.cc | 43 se::CudaComputeCapability GetComputeCapability() { in GetComputeCapability()
347 if (GetComputeCapability().IsAtLeast(se::CudaComputeCapability::PASCAL_)) { in TEST_F()
369 if (!GetComputeCapability().IsAtLeast(se::CudaComputeCapability::PASCAL_)) { in TEST_F()
|
| D | gpu_device.cc | 1032 se::CudaComputeCapability cc = in SingleVirtualDeviceMemoryLimit()
1036 !cc.IsAtLeast(se::CudaComputeCapability::PASCAL_)) { in SingleVirtualDeviceMemoryLimit()
1694 se::CudaComputeCapability ComputeCapabilityFromString( in ComputeCapabilityFromString()
1706 return se::CudaComputeCapability{major_part, minor_part}; in ComputeCapabilityFromString()
1709 std::vector<se::CudaComputeCapability> GetSupportedCudaComputeCapabilities() { in GetSupportedCudaComputeCapabilities()
1710 std::vector<se::CudaComputeCapability> cuda_caps = { in GetSupportedCudaComputeCapabilities()
1846 se::CudaComputeCapability min_supported_capability = *std::min_element( in GetValidDeviceIds()
|
| /external/tensorflow/tensorflow/compiler/xla/service/ |
| D | platform_util.cc | 134 se::CudaComputeCapability cc = description.cuda_compute_capability(); in IsDeviceSupported()
|
| /external/tensorflow/tensorflow/compiler/xla/tests/ |
| D | llvm_compiler_test.cc | 57 return se::CudaComputeCapability{0, 0}; in GetGpuVersion()
|
| /external/tensorflow/tensorflow/compiler/xla/service/gpu/llvm_gpu_backend/ |
| D | gpu_backend_lib.cc | 90 static string GetSmName(se::CudaComputeCapability compute_capability) { in GetSmName()
361 llvm::Triple target_triple, se::CudaComputeCapability compute_capability, in NVPTXGetTargetMachine()
535 absl::get_if<se::CudaComputeCapability>(&gpu_version); in CompileToPtx()
|
| /external/tensorflow/tensorflow/compiler/mlir/tools/kernel_gen/transforms/ |
| D | gpu_kernel_to_blob_pass.cc | 177 tensorflow::se::CudaComputeCapability{cc_major, cc_minor}, config, in GetGpuBinaryBlob()
|