Home
last modified time | relevance | path

Searched refs:blockIdx (Results 1 – 25 of 32) sorted by relevance

12

/external/clang/test/SemaCUDA/
Dcuda-builtin-vars.cu14 out[i++] = blockIdx.x; in kernel()
15 blockIdx.x = 0; // expected-error {{no setter defined for property 'x'}} in kernel()
16 out[i++] = blockIdx.y; in kernel()
17 blockIdx.y = 0; // expected-error {{no setter defined for property 'y'}} in kernel()
18 out[i++] = blockIdx.z; in kernel()
19 blockIdx.z = 0; // expected-error {{no setter defined for property 'z'}} in kernel()
/external/clang/include/clang/StaticAnalyzer/Core/PathSensitive/
DWorkList.h32 unsigned blockIdx; // This is the index of the next statement. variable
40 blockIdx(idx) {} in WorkListUnit()
46 blockIdx(0) {} in WorkListUnit()
58 unsigned getIndex() const { return blockIdx; } in getIndex()
/external/clang/test/CodeGenCUDA/
Dcuda-builtin-vars.cu13 out[i++] = blockIdx.x; // CHECK: call i32 @llvm.nvvm.read.ptx.sreg.ctaid.x() in kernel()
14 out[i++] = blockIdx.y; // CHECK: call i32 @llvm.nvvm.read.ptx.sreg.ctaid.y() in kernel()
15 out[i++] = blockIdx.z; // CHECK: call i32 @llvm.nvvm.read.ptx.sreg.ctaid.z() in kernel()
/external/tensorflow/tensorflow/core/kernels/
Dconcat_lib_gpu_impl.cu.cc40 IntType gidx = blockIdx.x * blockDim.x + threadIdx.x; in concat_fixed_kernel()
43 IntType gidy = blockIdx.y * blockDim.y + threadIdx.y; in concat_fixed_kernel()
68 IntType gidx = blockIdx.x * blockDim.x + threadIdx.x; in concat_variable_kernel()
107 IntType gidy = blockIdx.y * blockDim.y + threadIdx.y; in concat_variable_kernel()
Dbias_op_gpu.cu.cc126 for (int32 index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads; in BiasGradNHWC_SharedAtomics()
154 int32 bias_index = blockIdx.x % bias_size; in BiasGradNCHW_SharedAtomics()
155 int32 group_index = blockIdx.x / bias_size; in BiasGradNCHW_SharedAtomics()
Dreduction_gpu_kernels.cu.h180 const int bid = blockIdx.x;
218 const int row = (blockIdx.x * blockDim.x + threadIdx.x) / 32;
222 int gid = threadIdx.x + blockIdx.x * blockDim.x;
260 rows_per_warp * (blockIdx.y * blockDim.y + threadIdx.y);
301 out[col * gridDim.y + blockIdx.y] = s;
311 int row = blockIdx.y * blockDim.y + threadIdx.y;
312 int col = blockIdx.x * 32 + threadIdx.x;
347 min(blockDim.y, num_rows - blockIdx.y * blockDim.y);
353 out[col * gridDim.y + blockIdx.y] = s;
365 const int tid = threadIdx.x + blockIdx.x * blockDim.x;
[all …]
Drandom_op_gpu.cu.cc134 const int32 thread_id = blockIdx.x * blockDim.x + threadIdx.x; in Run()
175 const int32 thread_id = blockIdx.x * blockDim.x + threadIdx.x; in Run()
Dcheck_numerics_op_gpu.cu.cc39 const int32 thread_id = blockIdx.x * blockDim.x + threadIdx.x; in CheckNumericsKernel()
Dsplit_lib_gpu.cu.cc118 IntType gidx = blockIdx.x * blockDim.x + threadIdx.x; in split_v_kernel()
157 IntType gidy = blockIdx.y * blockDim.y + threadIdx.y; in split_v_kernel()
Dadjust_hsv_gpu.cu.h103 const int64 idx = (blockDim.x * blockIdx.x + threadIdx.x) * 3; in adjust_hsv_nhwc()
Dsoftmax_op_gpu.cu.cc41 const int tid = blockIdx.x * blockDim.x + threadIdx.x; in GenerateNormalizedProb()
Dparameterized_truncated_normal_op_gpu.cu.cc59 const int32 initial_offset = blockIdx.x * blockDim.x + threadIdx.x; in TruncatedNormalKernel()
/external/eigen/unsupported/Eigen/CXX11/src/Tensor/
DTensorReductionCuda.h115 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x; in ReductionInitKernel()
129 const Index first_index = blockIdx.x * BlockSize * NumPerThread + threadIdx.x; in FullReductionKernel()
205 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x; in ReductionInitKernelHalfFloat()
223 const Index first_index = blockIdx.x * BlockSize * NumPerThread + 2*threadIdx.x; in FullReductionKernelHalfFloat()
390 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x;
400 for (Index i = blockIdx.x; i < num_input_blocks; i += gridDim.x) {
461 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x;
476 for (Index i = blockIdx.x; i < num_input_blocks; i += gridDim.x) {
670 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x;
DTensorConvolution.h579 const int first_x = blockIdx.x * maxX;
584 const int first_plane = blockIdx.y * blockDim.y;
628 const int first_x = blockIdx.x * maxX;
633 const int first_y = blockIdx.y * maxY;
638 const int first_plane = blockIdx.z * blockDim.z;
697 const int first_x = blockIdx.x * maxX;
701 const int first_y = blockIdx.y * maxY;
705 const int first_z = blockIdx.z * maxZ;
DTensorRandom.h24 blockIdx.x * blockDim.x + threadIdx.x + in get_random_seed()
25 gridDim.x * blockDim.x * (blockIdx.y * blockDim.y + threadIdx.y); in get_random_seed()
DTensorContractionCuda.h26 const Index m_block_idx = blockIdx.x; in EigenContractionKernelInternal()
27 const Index n_block_idx = blockIdx.y; in EigenContractionKernelInternal()
509 const Index m_block_idx = blockIdx.x; in EigenContractionKernel()
510 const Index n_block_idx = blockIdx.y; in EigenContractionKernel()
1143 const Index m_block_idx = blockIdx.x; in EigenFloatContractionKernel()
1144 const Index n_block_idx = blockIdx.y; in EigenFloatContractionKernel()
1183 const Index m_block_idx = blockIdx.x; in EigenFloatContractionKernel16x16()
1184 const Index n_block_idx = blockIdx.y; in EigenFloatContractionKernel16x16()
/external/tensorflow/tensorflow/contrib/rnn/kernels/
Dlstm_ops_gpu.cu.cc48 const int batch_id = blockIdx.x * blockDim.x + threadIdx.x; in lstm_gates()
49 const int act_id = blockIdx.y * blockDim.y + threadIdx.y; in lstm_gates()
160 const int gid = blockDim.x * blockIdx.x + threadIdx.x; in concat_xh()
254 const int batch_id = blockIdx.x * blockDim.x + threadIdx.x; in lstm_gates_bprop()
255 const int act_id = blockIdx.y * blockDim.y + threadIdx.y; in lstm_gates_bprop()
/external/eigen/test/
Dcuda_common.h11 dim3 threadIdx, blockDim, blockIdx; variable
26 int i = threadIdx.x + blockIdx.x*blockDim.x; in run_on_cuda_meta_kernel()
/external/tensorflow/tensorflow/examples/adding_an_op/
Dcuda_op_kernel.cu.cc21 for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < N; in AddOneKernel()
/external/tensorflow/tensorflow/tools/ci_build/builds/user_ops/
Dcuda_op_kernel.cu.cc21 for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < N; in AddOneKernel()
/external/tensorflow/tensorflow/compiler/xla/service/gpu/llvm_gpu_backend/tests_data/
Dsaxpy.ll7 @blockIdx = external addrspace(1) global %struct.uint3
35 …truct.uint3, %struct.uint3* addrspacecast (%struct.uint3 addrspace(1)* @blockIdx to %struct.uint3*…
87 …truct.uint3, %struct.uint3* addrspacecast (%struct.uint3 addrspace(1)* @blockIdx to %struct.uint3*…
/external/clang/lib/Headers/
Dcuda_builtin_vars.h113 __CUDA_BUILTIN_VAR __cuda_builtin_blockIdx_t blockIdx; variable
/external/tensorflow/tensorflow/contrib/mpi_collectives/kernels/
Dring.cu.cc90 for (size_t i = blockIdx.x * blockDim.x + threadIdx.x; i < N; in elemwise_accum()
/external/tensorflow/tensorflow/core/util/
Dcuda_device_functions.h90 return detail::CudaGridRange<T>(blockIdx.x * blockDim.x + threadIdx.x, in CudaGridRangeX()
98 return detail::CudaGridRange<T>(blockIdx.y * blockDim.y + threadIdx.y, in CudaGridRangeY()
106 return detail::CudaGridRange<T>(blockIdx.z * blockDim.z + threadIdx.z, in CudaGridRangeZ()
/external/spirv-llvm/lib/SPIRV/
DOCLUtil.cpp365 getBlockInvokeTy(Function * F, unsigned blockIdx) { in getBlockInvokeTy() argument
367 PointerType * funcPtr = cast<PointerType>(params[blockIdx]); in getBlockInvokeTy()

12