Home
last modified time | relevance | path

Searched refs:NumThreads (Results 1 – 25 of 127) sorted by relevance

123456

/external/llvm-project/openmp/libomptarget/deviceRTLs/common/src/
Dreduction.cu84 uint32_t NumThreads = GetNumberOfOmpThreads(isSPMDExecutionMode); in nvptx_parallel_reduce_nowait() local
85 if (NumThreads == 1) in nvptx_parallel_reduce_nowait()
99 uint32_t WarpsNeeded = (NumThreads + WARPSIZE - 1) / WARPSIZE; in nvptx_parallel_reduce_nowait()
106 if ((NumThreads % WARPSIZE == 0) || (WarpId < WarpsNeeded - 1)) in nvptx_parallel_reduce_nowait()
108 else if (NumThreads > 1) // Only SPMD execution mode comes thru this case. in nvptx_parallel_reduce_nowait()
110 /*LaneCount=*/NumThreads % WARPSIZE, in nvptx_parallel_reduce_nowait()
117 if (NumThreads > WARPSIZE) { in nvptx_parallel_reduce_nowait()
144 if (NumThreads > WARPSIZE) { in nvptx_parallel_reduce_nowait()
145 uint32_t WarpsNeeded = (NumThreads + WARPSIZE - 1) / WARPSIZE; in nvptx_parallel_reduce_nowait()
204 uint32_t NumThreads = in __kmpc_nvptx_teams_reduce_nowait_v2() local
[all …]
Dparallel.cu55 uint16_t NumThreads = ThreadsAvailable; in determineNumberOfThreads() local
56 if (ThreadsRequested != 0 && ThreadsRequested < NumThreads) { in determineNumberOfThreads()
57 NumThreads = ThreadsRequested; in determineNumberOfThreads()
64 if (NumThreads < WARPSIZE) { in determineNumberOfThreads()
65 NumThreads = 1; in determineNumberOfThreads()
67 NumThreads = (NumThreads & ~((uint16_t)WARPSIZE - 1)); in determineNumberOfThreads()
71 return NumThreads; in determineNumberOfThreads()
97 uint16_t NumThreads = in __kmpc_kernel_prepare_parallel() local
105 ASSERT(LT_FUSSY, NumThreads > 0, "bad thread request of %d threads", in __kmpc_kernel_prepare_parallel()
106 (int)NumThreads); in __kmpc_kernel_prepare_parallel()
[all …]
/external/llvm-project/compiler-rt/lib/gwp_asan/tests/
Dthread_contention.cpp45 void runThreadContentionTest(unsigned NumThreads, unsigned NumIterations, in runThreadContentionTest() argument
50 if (std::thread::hardware_concurrency() < NumThreads) { in runThreadContentionTest()
51 NumThreads = std::thread::hardware_concurrency(); in runThreadContentionTest()
54 for (unsigned i = 0; i < NumThreads; ++i) { in runThreadContentionTest()
65 unsigned NumThreads = 4; in TEST_F() local
67 InitNumSlots(NumThreads); in TEST_F()
68 runThreadContentionTest(NumThreads, NumIterations, &GPA); in TEST_F()
Dmutex_test.cpp66 static void runSynchronisedTest(unsigned NumThreads, unsigned CounterMax) { in runSynchronisedTest() argument
69 ASSERT_TRUE(CounterMax % NumThreads == 0); in runSynchronisedTest()
75 for (unsigned i = 0; i < NumThreads; ++i) in runSynchronisedTest()
77 CounterMax / NumThreads); in runSynchronisedTest()
/external/gwp_asan/gwp_asan/tests/
Dthread_contention.cpp45 void runThreadContentionTest(unsigned NumThreads, unsigned NumIterations, in runThreadContentionTest() argument
50 if (std::thread::hardware_concurrency() < NumThreads) { in runThreadContentionTest()
51 NumThreads = std::thread::hardware_concurrency(); in runThreadContentionTest()
54 for (unsigned i = 0; i < NumThreads; ++i) { in runThreadContentionTest()
65 unsigned NumThreads = 4; in TEST_F() local
67 InitNumSlots(NumThreads); in TEST_F()
68 runThreadContentionTest(NumThreads, NumIterations, &GPA); in TEST_F()
Dmutex_test.cpp66 static void runSynchronisedTest(unsigned NumThreads, unsigned CounterMax) { in runSynchronisedTest() argument
69 ASSERT_TRUE(CounterMax % NumThreads == 0); in runSynchronisedTest()
75 for (unsigned i = 0; i < NumThreads; ++i) in runSynchronisedTest()
77 CounterMax / NumThreads); in runSynchronisedTest()
/external/llvm-project/openmp/libomptarget/deviceRTLs/nvptx/test/parallel/
Dlevel.c7 const int NumThreads = 64; variable
45 #pragma omp parallel num_threads(NumThreads) in main()
98 if (i < NumThreads) { in main()
108 if (i < NumThreads) { in main()
119 if (i < NumThreads) { in main()
129 if (i < NumThreads) { in main()
Dnested.c7 const int NumThreads = 64; variable
25 #pragma omp parallel num_threads(NumThreads) in main()
53 int Expected = NumThreads + 1; in main()
54 if (i < NumThreads) { in main()
64 if (i < NumThreads) { in main()
/external/swiftshader/third_party/subzero/src/
DLinuxMallocProfiling.cpp67 LinuxMallocProfiling::LinuxMallocProfiling(size_t NumThreads, Ostream *Ls) in LinuxMallocProfiling() argument
69 if (NumThreads != 0) { in LinuxMallocProfiling()
102 LinuxMallocProfiling::LinuxMallocProfiling(size_t NumThreads, Ostream *Ls) { in LinuxMallocProfiling() argument
103 (void)NumThreads; in LinuxMallocProfiling()
DIceBrowserCompileServer.cpp116 char *onInitCallback(uint32_t NumThreads, int *ObjFileFDs, in onInitCallback() argument
152 gCompileServer->getParsedFlags(UseNumThreadsFromBrowser, NumThreads, in onInitCallback()
242 uint32_t NumThreads, int argc, in getParsedFlags() argument
248 ClFlags::Flags.setNumTranslationThreads(NumThreads); in getParsedFlags()
DLinuxMallocProfiling.h32 LinuxMallocProfiling(size_t NumThreads, Ostream *Ls);
/external/swiftshader/third_party/subzero/crosstest/
Dtest_sync_atomic_main.cpp241 const size_t NumThreads = 4; in testAtomicRMWThreads() local
242 pthread_t t[NumThreads]; in testAtomicRMWThreads()
243 pthread_attr_t attr[NumThreads]; in testAtomicRMWThreads()
247 for (size_t m = 0; m < NumThreads; ++m) { in testAtomicRMWThreads()
255 for (size_t m = 0; m < NumThreads; ++m) { in testAtomicRMWThreads()
262 for (size_t m = 0; m < NumThreads; ++m) { in testAtomicRMWThreads()
273 for (size_t m = 0; m < NumThreads; ++m) { in testAtomicRMWThreads()
/external/tensorflow/tensorflow/core/common_runtime/
Dprocess_util_test.cc22 TEST(ProcessUtilTest, NumThreads) { in TEST() argument
33 EXPECT_EQ(10, pool->NumThreads()); in TEST()
Drenamed_device.cc62 eigen_worker_threads_.num_threads = underlying_threadpool->NumThreads(); in RenamedDevice()
67 underlying_threadpool, underlying_threadpool->NumThreads()); in RenamedDevice()
/external/llvm-project/clang-tools-extra/clangd/unittests/support/
DThreadingTests.cpp67 const unsigned NumThreads = 5; in TEST_F() local
77 for (unsigned I = 0; I < NumThreads; ++I) in TEST_F()
89 EXPECT_LE(ComputeCount, NumThreads * NumKeys) in TEST_F()
/external/tensorflow/tensorflow/core/platform/
Dthreadpool.cc111 underlying_threadpool_, underlying_threadpool_->NumThreads(), nullptr)); in ThreadPool()
124 NumThreads() == 1) { in NumShardsUsedByFixedBlockSizeScheduling()
191 if (num_shards_used <= NumThreads()) { in ParallelForFixedBlockSizeScheduling()
243 int ThreadPool::NumThreads() const { in NumThreads() function in tensorflow::thread::ThreadPool
244 return underlying_threadpool_->NumThreads(); in NumThreads()
/external/lzma/CPP/7zip/Archive/7z/
D7zCompressionMode.h55 UInt32 NumThreads; member
68 , NumThreads(1) in CCompressionMethodMode()
/external/tensorflow/tensorflow/compiler/aot/
Dtest.cc63 Eigen::ThreadPoolDevice device(&pool, pool.NumThreads()); in TEST()
77 Eigen::ThreadPoolDevice device(&pool, pool.NumThreads()); in BM_NAME()
/external/lzma/CPP/7zip/UI/Common/
DZipRegistry.h44 UInt32 NumThreads; member
58 BlockLogSize = NumThreads = Level = Dictionary = Order = (UInt32)(Int32)-1; in ResetForLevelChange()
/external/tensorflow/tensorflow/compiler/aot/tests/
Dtfcompile_test.cc247 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
318 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
344 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
450 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
478 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
502 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
516 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
538 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
565 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
661 Eigen::ThreadPoolDevice device(&tp, tp.NumThreads()); in TEST()
/external/llvm-project/llvm/examples/SpeculativeJIT/
DSpeculativeJIT.cpp34 static cl::opt<unsigned> NumThreads("num-threads", cl::Optional, variable
137 ThreadPool CompileThreads{llvm::hardware_concurrency(NumThreads)};
161 if (NumThreads < 1) { in main()
/external/tensorflow/tensorflow/core/kernels/
Dconv_2d_gpu.h274 template <typename T, int NumThreads, int TileSizeI, int TileSizeJ,
279 eigen_assert(blockDim.x == NumThreads);
285 constexpr int ReadRowPerPass = NumThreads / TileSizeJ;
286 constexpr int WriteRowPerPass = NumThreads / TileSizeI;
345 constexpr int in_effective_thread_num = NumThreads / TileSizeJ * TileSizeJ;
391 constexpr int out_effective_thread_num = NumThreads / TileSizeI * TileSizeI;
702 constexpr int NumThreads = TileLongSide;
705 SwapDimension1And2InTensor3UsingTiles<T, NumThreads, TileLongSide,
707 total_tiles_count, NumThreads, 0, d.stream(), input, input_dims,
711 SwapDimension1And2InTensor3UsingTiles<T, NumThreads, TileShortSide,
[all …]
/external/swiftshader/third_party/marl/tools/bench/
Dbench.go32 NumThreads uint member
50 t.NumThreads = uint(n)
/external/tensorflow/tensorflow/lite/kernels/
Deigen_support.cc83 int NumThreads() const override { return pool_ ? pool_->NumThreads() : 1; } in NumThreads() function in tflite::eigen_support::__anonf8036f140111::EigenThreadPoolWrapper
/external/eigen/unsupported/Eigen/CXX11/src/ThreadPool/
DThreadPoolInterface.h22 virtual int NumThreads() const = 0;

123456