/external/tensorflow/tensorflow/compiler/xla/service/gpu/tests/ |
D | reduction_vectorization_test.cc | 56 int cc_major = 0, cc_minor = 0; in TEST_F() local 58 &cc_minor); in TEST_F() 104 int cc_major = 0, cc_minor = 0; in TEST_F() local 106 &cc_minor); in TEST_F() 152 int cc_major = 0, cc_minor = 0; in TEST_F() local 154 &cc_minor); in TEST_F() 236 int cc_major = 0, cc_minor = 0; in TEST_F() local 238 &cc_minor); in TEST_F() 341 int cc_major = 0, cc_minor = 0; in TEST_F() local 342 device_description.cuda_compute_capability(&cc_major, &cc_minor); in TEST_F()
|
D | gpu_atomic_test.cc | 114 int cc_major = 0, cc_minor = 0; in TEST_F() local 115 device_description.cuda_compute_capability(&cc_major, &cc_minor); in TEST_F()
|
D | hlo_to_llvm_ir.cc | 55 cuda_compute_capability.cc_minor = 0; in CompileAndPrintLlvmIr()
|
D | mlir_gpu_test_base.cc | 60 &cuda_compute_capability.cc_major, &cuda_compute_capability.cc_minor); in CompileMlirModule()
|
/external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
D | nvptx_compiler.h | 79 int cc_minor, const HloModuleConfig& hlo_module_config, bool relocatable); 94 CompilationCacheKey(std::string ptx, int cc_major, int cc_minor, in CompilationCacheKey() 98 cc_minor(cc_minor), in CompilationCacheKey() 102 int cc_minor; member 111 key.cc_minor), in operator() 118 return a.cc_major == b.cc_major && a.cc_minor == b.cc_minor && in operator()
|
D | nvptx_compiler.cc | 288 int cc_major, cc_minor; in GetGpuVersion() local 290 &cc_minor)) { in GetGpuVersion() 294 cc_minor = 0; in GetGpuVersion() 297 return std::make_pair(cc_major, cc_minor); in GetGpuVersion() 343 int cc_minor, const HloModuleConfig& hlo_module_config, bool relocatable) { in CompileGpuAsmOrGetCachedResult() argument 358 std::forward_as_tuple(ptx, cc_major, cc_minor, relocatable), in CompileGpuAsmOrGetCachedResult()
|
D | gpu_device_info.h | 28 int cc_minor; member
|
D | gpu_conv_algorithm_picker.cc | 181 int cc_major, cc_minor; in GetComputeCapability() local 183 &cc_minor); in GetComputeCapability() 185 cc.set_minor(cc_minor); in GetComputeCapability()
|
/external/tensorflow/tensorflow/core/grappler/ |
D | devices.cc | 53 int cc_minor = 0; in GetNumAvailableGPUs() local 54 desc->cuda_compute_capability(&cc_major, &cc_minor); in GetNumAvailableGPUs() 55 std::pair<int, int> cuda_compute_capability(cc_major, cc_minor); in GetNumAvailableGPUs()
|
/external/tensorflow/tensorflow/stream_executor/gpu/ |
D | asm_compiler.cc | 147 int cc_minor; in CompileGpuAsm() local 149 gpu::GpuDriver::GetComputeCapability(&cc_major, &cc_minor, handle)); in CompileGpuAsm() 150 return CompileGpuAsm(cc_major, cc_minor, ptx_contents, options); in CompileGpuAsm() 181 int cc_minor) { in LogPtxasTooOld() argument 190 if (already_logged->insert({ptxas_path, cc_major, cc_minor}).second) { in LogPtxasTooOld() 193 << cc_major << "." << cc_minor; in LogPtxasTooOld() 198 port::StatusOr<std::vector<uint8>> CompileGpuAsm(int cc_major, int cc_minor, in CompileGpuAsm() argument 233 absl::StrCat("-arch=sm_", cc_major, cc_minor)}; in CompileGpuAsm() 263 LogPtxasTooOld(ptxas_path, cc_major, cc_minor); in CompileGpuAsm()
|
D | asm_compiler.h | 47 port::StatusOr<std::vector<uint8>> CompileGpuAsm(int cc_major, int cc_minor,
|
D | gpu_driver.h | 464 static port::Status GetComputeCapability(int* cc_major, int* cc_minor,
|
/external/tensorflow/tensorflow/core/common_runtime/gpu/ |
D | gpu_device_test.cc | 44 int* cc_minor) { in GetComputeCapability() argument 48 if (!se->GetDeviceDescription().cuda_compute_capability(cc_major, cc_minor)) { in GetComputeCapability() 50 *cc_minor = 0; in GetComputeCapability() 352 int cc_major, cc_minor; in TEST_F() local 353 TF_ASSERT_OK(GetComputeCapability(PlatformGpuId(0), &cc_major, &cc_minor)); in TEST_F() 376 int cc_major, cc_minor; in TEST_F() local 377 TF_ASSERT_OK(GetComputeCapability(kPlatformGpuId, &cc_major, &cc_minor)); in TEST_F()
|
D | gpu_device.cc | 971 int cc_major = 0, cc_minor = 0; in SingleVirtualDeviceMemoryLimit() local 973 &cc_minor)) { in SingleVirtualDeviceMemoryLimit() 1135 int cc_major, cc_minor; in GetDeviceDetails() local 1136 if (desc->cuda_compute_capability(&cc_major, &cc_minor)) { in GetDeviceDetails() 1137 (*details)["compute_capability"] = strings::StrCat(cc_major, ".", cc_minor); in GetDeviceDetails() 1383 int cc_minor; in GetShortDeviceDescription() local 1384 if (!desc.cuda_compute_capability(&cc_major, &cc_minor)) { in GetShortDeviceDescription() 1386 cc_minor = 0; in GetShortDeviceDescription() 1392 ", compute capability: ", cc_major, ".", cc_minor); in GetShortDeviceDescription() 1764 int cc_minor; in GetValidDeviceIds() local [all …]
|
/external/tensorflow/tensorflow/compiler/mlir/tools/kernel_gen/transforms/ |
D | gpu_kernel_to_blob_pass.cc | 185 uint32_t cc_minor = arch % 10; in GetGpuBinaryBlob() local 191 std::make_pair(cc_major, cc_minor), in GetGpuBinaryBlob() 203 cc_major, cc_minor, ptx.c_str(), gpu_asm_opts)); in GetGpuBinaryBlob()
|
/external/tensorflow/tensorflow/stream_executor/ |
D | stream_executor_pimpl.cc | 271 int cc_major, cc_minor; in GetConvolveAlgorithms() local 272 GetDeviceDescription().cuda_compute_capability(&cc_major, &cc_minor); in GetConvolveAlgorithms() 274 cc_minor, out_algorithms); in GetConvolveAlgorithms() 312 int cc_major, cc_minor; in GetConvolveBackwardDataAlgorithms() local 313 GetDeviceDescription().cuda_compute_capability(&cc_major, &cc_minor); in GetConvolveBackwardDataAlgorithms() 315 with_winograd_nonfused, cc_major, cc_minor, out_algorithms); in GetConvolveBackwardDataAlgorithms() 325 int cc_major, cc_minor; in GetConvolveBackwardFilterAlgorithms() local 326 GetDeviceDescription().cuda_compute_capability(&cc_major, &cc_minor); in GetConvolveBackwardFilterAlgorithms() 328 with_winograd_nonfused, cc_major, cc_minor, out_algorithms); in GetConvolveBackwardFilterAlgorithms()
|
D | dnn.cc | 45 bool with_winograd_nonfused, int cc_major, int cc_minor, in GetConvolveAlgorithms() argument 69 bool with_winograd_nonfused, int cc_major, int cc_minor, in GetConvolveBackwardDataAlgorithms() argument 75 bool with_winograd_nonfused, int cc_major, int cc_minor, in GetConvolveBackwardFilterAlgorithms() argument
|
/external/tensorflow/tensorflow/core/kernels/ |
D | gpu_utils.cc | 116 int cc_major, cc_minor; in GetComputeCapability() local 118 &cc_minor); in GetComputeCapability() 120 cc.set_minor(cc_minor); in GetComputeCapability()
|
/external/tensorflow/tensorflow/compiler/mlir/tensorflow/utils/ |
D | device_util_test.cc | 94 ASSERT_EQ(device_meta_1.cc_minor().getInt(), 0); in TEST() 202 ASSERT_EQ(meta_0->cc_minor().getInt(), 2); in TEST()
|
/external/tensorflow/tensorflow/stream_executor/cuda/ |
D | cuda_gpu_executor.cc | 970 int cc_minor; in CreateDeviceDescription() local 971 status = GpuDriver::GetComputeCapability(&cc_major, &cc_minor, device); in CreateDeviceDescription() 1060 absl::StrCat("Compute Capability ", cc_major, ".", cc_minor)); in CreateDeviceDescription() 1067 builder.set_cuda_compute_capability(cc_major, cc_minor); in CreateDeviceDescription()
|
D | cuda_dnn.cc | 672 int cc_major, cc_minor; in GetCcMajorMinor() local 674 &cc_minor); in GetCcMajorMinor() 675 return std::make_tuple(cc_major, cc_minor); in GetCcMajorMinor() 758 int cc_major, cc_minor; in IsTensorMathEnabled() local 759 std::tie(cc_major, cc_minor) = GetCcMajorMinor(stream); in IsTensorMathEnabled() 3295 bool with_winograd_nonfused, int cc_major, int cc_minor, in GetConvolveAlgorithms() argument 3365 bool with_winograd_nonfused, int cc_major, int cc_minor, in GetConvolveBackwardDataAlgorithms() argument 3404 bool with_winograd_nonfused, int cc_major, int cc_minor, in GetConvolveBackwardFilterAlgorithms() argument 3836 int cc_major, cc_minor; in DoFusedConvolve() local 3837 std::tie(cc_major, cc_minor) = GetCcMajorMinor(stream); in DoFusedConvolve() [all …]
|
D | cuda_dnn.h | 204 bool with_winograd_nonfused, int cc_major, int cc_minor, 211 bool with_winograd_nonfused, int cc_major, int cc_minor, 215 bool with_winograd_nonfused, int cc_major, int cc_minor,
|
/external/tensorflow/tensorflow/compiler/mlir/tensorflow/tests/ |
D | layout_optimization_layout_assignment_gpu_cc_60.mlir | 4 tf.devices = {"/device:GPU:0" = {cc_major = 6 : i32, cc_minor = 0 : i32}}
|
D | layout_optimization_layout_assignment_gpu_cc_70.mlir | 4 tf.devices = {"/device:GPU:0" = {cc_major = 7 : i32, cc_minor = 0 : i32}}
|
/external/tensorflow/tensorflow/stream_executor/rocm/ |
D | rocm_dnn.h | 231 bool with_winograd_nonfused, int cc_major, int cc_minor, 249 bool with_winograd_nonfused, int cc_major, int cc_minor, 253 bool with_winograd_nonfused, int cc_major, int cc_minor,
|