| /libCEED/backends/hip-shared/ |
| H A D | ceed-hip-shared-basis.c | 44 const CeedInt thread_1d = CeedIntMax(P_1d, Q_1d); in ComputeBasisThreadBlockSizes() 61 block_sizes[0] = CeedIntMax(256, ComputeBlockSizeFromRequirement(required)); in ComputeBasisThreadBlockSizes() 64 block_sizes[1] = CeedIntMax(256, ComputeBlockSizeFromRequirement(required)); in ComputeBasisThreadBlockSizes() 67 required = CeedIntMax(64, Q_1d * Q_1d); in ComputeBasisThreadBlockSizes() 68 block_sizes[2] = CeedIntMax(256, ComputeBlockSizeFromRequirement(required)); in ComputeBasisThreadBlockSizes() 75 block_sizes[0] = CeedIntMax(256, ComputeBlockSizeFromRequirement(required)); in ComputeBasisThreadBlockSizes() 78 block_sizes[1] = CeedIntMax(256, ComputeBlockSizeFromRequirement(required)); in ComputeBasisThreadBlockSizes() 82 block_sizes[2] = CeedIntMax(256, ComputeBlockSizeFromRequirement(required)); in ComputeBasisThreadBlockSizes() 124 CeedInt thread_1d = CeedIntMax(Q_1d, P_1d); in CeedBasisApplyTensorCore_Hip_shared() 141 const CeedInt elems_per_block = CeedIntMax(block_size / (thread_1d * thread_1d), 1); in CeedBasisApplyTensorCore_Hip_shared() [all …]
|
| /libCEED/backends/cuda-shared/ |
| H A D | ceed-cuda-shared-basis.c | 56 CeedInt thread_1d = CeedIntMax(Q_1d, P_1d); in CeedBasisApplyTensorCore_Cuda_shared() 62 …CeedInt elems_per_block = CeedIntMin(ceed_Cuda->device_prop.maxThreadsDim[2], CeedIntMax(512 / thr… in CeedBasisApplyTensorCore_Cuda_shared() 75 … CeedInt elems_per_block = CeedIntMax(thread_1d < 7 ? opt_elems[thread_1d] / num_comp : 1, 1); in CeedBasisApplyTensorCore_Cuda_shared() 106 CeedInt thread_1d = CeedIntMax(Q_1d, P_1d); in CeedBasisApplyTensorCore_Cuda_shared() 116 …CeedInt elems_per_block = CeedIntMin(ceed_Cuda->device_prop.maxThreadsDim[2], CeedIntMax(512 / thr… in CeedBasisApplyTensorCore_Cuda_shared() 129 … CeedInt elems_per_block = CeedIntMax(thread_1d < 7 ? opt_elems[thread_1d] / num_comp : 1, 1); in CeedBasisApplyTensorCore_Cuda_shared() 235 for (CeedInt i = 1; i < num_elem; i++) max_num_points = CeedIntMax(max_num_points, num_points[i]); in CeedBasisApplyAtPointsCore_Cuda_shared() 293 …CeedIntMax(Q_1d, P_1d), "BASIS_DIM", dim, "BASIS_NUM_COMP", num_comp, "BASIS_NUM_NODES", CeedIntPo… in CeedBasisApplyAtPointsCore_Cuda_shared() 320 CeedInt thread_1d = CeedIntMax(Q_1d, P_1d); in CeedBasisApplyAtPointsCore_Cuda_shared() 326 …CeedInt elems_per_block = CeedIntMin(ceed_Cuda->device_prop.maxThreadsDim[2], CeedIntMax(512 / thr… in CeedBasisApplyAtPointsCore_Cuda_shared() [all …]
|
| /libCEED/backends/cuda-ref/ |
| H A D | ceed-cuda-ref-qfunction-load.cpp | 66 code << " const CeedScalar *inputs[" << CeedIntMax(num_input_fields, 1) << "];\n"; in CeedQFunctionBuildKernel_Cuda_ref() 79 code << " CeedScalar *outputs[" << CeedIntMax(num_output_fields, 1) << "];\n"; in CeedQFunctionBuildKernel_Cuda_ref()
|
| H A D | ceed-cuda-ref-restriction.c | 519 max_points = CeedIntMax(max_points, offsets[i + 1] - offsets[i]); in CeedElemRestrictionCreate_Cuda()
|
| H A D | ceed-cuda-ref-basis.c | 125 for (CeedInt i = 1; i < num_elem; i++) max_num_points = CeedIntMax(max_num_points, num_points[i]); in CeedBasisApplyAtPointsCore_Cuda()
|
| /libCEED/backends/hip-ref/ |
| H A D | ceed-hip-ref-qfunction-load.cpp | 70 code << " const CeedScalar *inputs[" << CeedIntMax(num_input_fields, 1) << "];\n"; in CeedQFunctionBuildKernel_Hip_ref() 83 code << " CeedScalar *outputs[" << CeedIntMax(num_output_fields, 1) << "];\n"; in CeedQFunctionBuildKernel_Hip_ref()
|
| H A D | ceed-hip-ref-restriction.c | 520 max_points = CeedIntMax(max_points, offsets[i + 1] - offsets[i]); in CeedElemRestrictionCreate_Hip()
|
| H A D | ceed-hip-ref-basis.c | 124 for (CeedInt i = 1; i < num_elem; i++) max_num_points = CeedIntMax(max_num_points, num_points[i]); in CeedBasisApplyAtPointsCore_Hip()
|
| /libCEED/backends/sycl-ref/ |
| H A D | ceed-sycl-ref-qfunction-load.sycl.cpp | 121 code << " const CeedScalar *inputs[" << CeedIntMax(num_input_fields, 1) << "] = {U_0"; in CeedQFunctionBuildKernel_Sycl() 132 code << " CeedScalar *outputs[" << CeedIntMax(num_output_fields, 1) << "] = {V_0"; in CeedQFunctionBuildKernel_Sycl()
|
| H A D | ceed-sycl-ref-basis.sycl.cpp | 583 impl->buf_len = num_comp * CeedIntMax(num_nodes, num_qpts); in CeedBasisCreateTensorH1_Sycl()
|
| /libCEED/backends/cuda-gen/ |
| H A D | ceed-cuda-gen-operator-build.cpp | 84 *max_dim = CeedIntMax(*max_dim, field_dim); in CeedOperatorBuildKernelData_Cuda_gen() 88 *max_P = CeedIntMax(*max_P, field_P); in CeedOperatorBuildKernelData_Cuda_gen() 92 *max_P_1d = CeedIntMax(*max_P_1d, field_P_1d); in CeedOperatorBuildKernelData_Cuda_gen() 116 *max_dim = CeedIntMax(*max_dim, field_dim); in CeedOperatorBuildKernelData_Cuda_gen() 120 *max_P = CeedIntMax(*max_P, field_P); in CeedOperatorBuildKernelData_Cuda_gen() 124 *max_P_1d = CeedIntMax(*max_P_1d, field_P_1d); in CeedOperatorBuildKernelData_Cuda_gen() 984 code << tab << "CeedScalar *inputs[" << CeedIntMax(num_input_fields, 1) << "];\n"; in CeedOperatorBuildKernelQFunction_Cuda_gen() 993 code << tab << "CeedScalar *outputs[" << CeedIntMax(num_output_fields, 1) << "];\n"; in CeedOperatorBuildKernelQFunction_Cuda_gen() 1501 …max_rstr_buffer_size = CeedIntMax(max_rstr_buffer_size, num_comp * (is_all_tensor && (max_dim >= 3… in CeedOperatorBuildKernel_Cuda_gen() 1515 …max_rstr_buffer_size = CeedIntMax(max_rstr_buffer_size, num_comp * (is_all_tensor && (max_dim >= 3… in CeedOperatorBuildKernel_Cuda_gen() [all …]
|
| H A D | ceed-cuda-gen-operator.c | 215 …CeedInt elems_per_block = CeedIntMin(cuda_data->device_prop.maxThreadsDim[2], CeedIntMax(512 / dat… in CeedOperatorApplyAddCore_Cuda_gen() 470 …CeedInt elems_per_block = CeedIntMin(cuda_data->device_prop.maxThreadsDim[2], CeedIntMax(512 / dat… in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen()
|
| /libCEED/backends/magma/ |
| H A D | ceed-magma-basis.c | 92 CeedInt max_P_Q = CeedIntMax(P, Q); in CeedBasisApplyCore_Magma() 113 shared_mem += sizeof(CeedScalar) * num_t_col * (CeedIntMax(P * P * max_P_Q, P * Q * Q)); in CeedBasisApplyCore_Magma() 169 CeedInt max_P_Q = CeedIntMax(P, Q); in CeedBasisApplyCore_Magma() 190 … shared_mem += sizeof(CeedScalar) * num_t_col * CeedIntMax(P * P * P, (P * P * Q) + (P * Q * Q)); in CeedBasisApplyCore_Magma() 429 …ode != CEED_TRANSPOSE && q_comp > 1) ? (shared_mem_A + shared_mem_B) : CeedIntMax(shared_mem_A, sh… in CeedBasisApplyNonTensorCore_Magma() 591 P_1d, "BASIS_Q", Q_1d, "BASIS_MAX_P_Q", CeedIntMax(P_1d, Q_1d))); in CeedBasisCreateTensorH1_Magma()
|
| /libCEED/backends/hip-gen/ |
| H A D | ceed-hip-gen-operator-build.cpp | 35 const CeedInt thread_1d = CeedIntMax(Q_1d, P_1d); in BlockGridCalculate_Hip_gen() 111 *max_dim = CeedIntMax(*max_dim, field_dim); in CeedOperatorBuildKernelData_Hip_gen() 115 *max_P = CeedIntMax(*max_P, field_P); in CeedOperatorBuildKernelData_Hip_gen() 119 *max_P_1d = CeedIntMax(*max_P_1d, field_P_1d); in CeedOperatorBuildKernelData_Hip_gen() 143 *max_dim = CeedIntMax(*max_dim, field_dim); in CeedOperatorBuildKernelData_Hip_gen() 147 *max_P = CeedIntMax(*max_P, field_P); in CeedOperatorBuildKernelData_Hip_gen() 151 *max_P_1d = CeedIntMax(*max_P_1d, field_P_1d); in CeedOperatorBuildKernelData_Hip_gen() 1011 code << tab << "CeedScalar *inputs[" << CeedIntMax(num_input_fields, 1) << "];\n"; in CeedOperatorBuildKernelQFunction_Hip_gen() 1020 code << tab << "CeedScalar *outputs[" << CeedIntMax(num_output_fields, 1) << "];\n"; in CeedOperatorBuildKernelQFunction_Hip_gen() 1514 …max_rstr_buffer_size = CeedIntMax(max_rstr_buffer_size, num_comp * (is_all_tensor && (max_dim >= 3… in CeedOperatorBuildKernel_Hip_gen() [all …]
|
| /libCEED/backends/sycl-shared/ |
| H A D | ceed-sycl-shared-basis.sycl.cpp | 182 const CeedInt thread_1d = CeedIntMax(Q_1d, P_1d); in CeedBasisCreateTensorH1_Sycl_shared()
|
| /libCEED/backends/sycl-gen/ |
| H A D | ceed-sycl-gen-operator-build.sycl.cpp | 31 const CeedInt thread1d = CeedIntMax(Q_1d, P_1d); in BlockGridCalculate_Sycl_gen()
|
| /libCEED/include/ceed/ |
| H A D | ceed.h | 559 static inline CeedInt CeedIntMax(CeedInt a, CeedInt b) { return a > b ? a : b; } in CeedIntMax() function
|
| /libCEED/backends/memcheck/ |
| H A D | ceed-memcheck-restriction.c | 683 max_points = CeedIntMax(max_points, num_points); in CeedElemRestrictionCreate_Memcheck()
|
| /libCEED/julia/LibCEED.jl/src/generated/ |
| H A D | libceed_bindings.jl | 857 function CeedIntMax(a, b) function 858 ccall((:CeedIntMax, libceed), CeedInt, (CeedInt, CeedInt), a, b)
|
| /libCEED/interface/ |
| H A D | ceed-elemrestriction.c | 1602 if (max_points) *max_points = CeedIntMax(num_points, *max_points); in CeedElemRestrictionGetMinMaxPointsInElement()
|
| /libCEED/backends/ref/ |
| H A D | ceed-ref-restriction.c | 788 max_points = CeedIntMax(max_points, num_points); in CeedElemRestrictionCreate_Ref()
|
| /libCEED/ |
| H A D | coverage.info | 4945 FNA:2,0,CeedIntMax
|