Home
last modified time | relevance | path

Searched defs:grid (Results 1 – 12 of 12) sorted by relevance

/libCEED/backends/cuda-shared/
H A Dceed-cuda-shared-basis.c63 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Cuda_shared() local
76 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Cuda_shared() local
88 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Cuda_shared() local
117 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Cuda_shared() local
130 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Cuda_shared() local
141 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Cuda_shared() local
327 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Cuda_shared() local
341 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Cuda_shared() local
353 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Cuda_shared() local
377 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Cuda_shared() local
[all …]
/libCEED/backends/hip-shared/
H A Dceed-hip-shared-basis.c130 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Hip_shared() local
142 … CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Hip_shared() local
153 … CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Hip_shared() local
182 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Hip_shared() local
194 … CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Hip_shared() local
205 … CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyTensorCore_Hip_shared() local
393 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Hip_shared() local
405 … CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Hip_shared() local
417 … CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Hip_shared() local
441 CeedInt grid = num_elem / elems_per_block + (num_elem % elems_per_block > 0); in CeedBasisApplyAtPointsCore_Hip_shared() local
[all …]
/libCEED/backends/hip-gen/
H A Dceed-hip-gen-operator.c163 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorApplyAddCore_Hip_gen() local
169 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorApplyAddCore_Hip_gen() local
175 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorApplyAddCore_Hip_gen() local
441 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorLinearAssembleQFunctionCore_Hip_gen() local
447 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorLinearAssembleQFunctionCore_Hip_gen() local
453 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorLinearAssembleQFunctionCore_Hip_gen() local
621 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() local
627 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() local
633 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() local
803 …CeedInt grid = num_elem / block_sizes[2] + ((num_elem / block_sizes[2] * block_sizes[2] < num… in CeedOperatorAssembleSingleAtPoints_Hip_gen() local
[all …]
/libCEED/backends/magma/
H A Dceed-magma-basis.c116 CeedInt grid = CeedDivUpInt(num_elem, num_t_col); in CeedBasisApplyCore_Magma() local
193 CeedInt grid = CeedDivUpInt(num_elem, num_t_col); in CeedBasisApplyCore_Magma() local
230 CeedInt grid = CeedDivUpInt(num_elem, num_t_col); in CeedBasisApplyCore_Magma() local
426 CeedInt grid = CeedDivUpInt(N, num_t_col * NB); in CeedBasisApplyNonTensorCore_Magma() local
447 CeedInt grid = CeedDivUpInt(num_elem, num_t_col); in CeedBasisApplyNonTensorCore_Magma() local
/libCEED/backends/cuda-gen/
H A Dceed-cuda-gen-operator.c76 int *grid) { in BlockGridCalculate()
205 int max_threads_per_block, min_grid_size, grid; in CeedOperatorApplyAddCore_Cuda_gen() local
460 int max_threads_per_block, min_grid_size, grid; in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen() local
636 int max_threads_per_block, min_grid_size, grid; in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen() local
801 int max_threads_per_block, min_grid_size, grid; in CeedOperatorAssembleSingleAtPoints_Cuda_gen() local
/libCEED/backends/sycl-gen/
H A Dceed-sycl-gen-operator.sycl.cpp126 CeedInt block_sizes[3], grid = 0; in CeedOperatorApplyAdd_Sycl_gen() local
/libCEED/backends/cuda-ref/
H A Dceed-cuda-ref-restriction.c150 const CeedInt grid = CeedDivUpInt(impl->num_nodes, block_size); in CeedElemRestrictionApply_Cuda_Core() local
195 const CeedInt grid = CeedDivUpInt(impl->num_nodes, block_size); in CeedElemRestrictionApply_Cuda_Core() local
H A Dceed-cuda-ref-basis.c261 const int grid = CeedDivUpInt(num_elem, elems_per_block); in CeedBasisApplyNonTensorCore_Cuda() local
H A Dceed-cuda-ref-operator.c1466 CeedInt grid = CeedDivUpInt(num_elem, elems_per_block); in CeedOperatorAssembleDiagonalCore_Cuda() local
1784 CeedInt grid = CeedDivUpInt(num_elem_in, asmb->elems_per_block); in CeedOperatorAssembleSingle_Cuda() local
/libCEED/backends/hip-ref/
H A Dceed-hip-ref-restriction.c151 const CeedInt grid = CeedDivUpInt(impl->num_nodes, block_size); in CeedElemRestrictionApply_Hip_Core() local
196 const CeedInt grid = CeedDivUpInt(impl->num_nodes, block_size); in CeedElemRestrictionApply_Hip_Core() local
H A Dceed-hip-ref-basis.c260 const int grid = CeedDivUpInt(num_elem, elems_per_block); in CeedBasisApplyNonTensorCore_Hip() local
H A Dceed-hip-ref-operator.c1463 CeedInt grid = CeedDivUpInt(num_elem, elems_per_block); in CeedOperatorAssembleDiagonalCore_Hip() local
1781 CeedInt grid = CeedDivUpInt(num_elem_in, asmb->elems_per_block); in CeedOperatorAssembleSingle_Hip() local