| /libCEED/include/ceed/jit-source/hip/ |
| H A D | hip-ref-restriction-at-points.h | 16 …AtPointsTranspose(const CeedInt *__restrict__ indices, const CeedInt *__restrict__ points_per_elem, in AtPointsTranspose() argument 23 if (loc_node >= points_per_elem[elem]) continue; in AtPointsTranspose() 31 … const CeedInt *__restrict__ points_per_elem, const CeedInt *__restrict__ t_offsets, in AtPointsTranspose() argument 47 if (loc_node >= points_per_elem[elem]) continue; in AtPointsTranspose()
|
| H A D | hip-shared-basis-tensor-at-points.h | 24 …Points(const CeedInt num_elem, const CeedScalar *c_B, const CeedInt *points_per_elem, const CeedSc… in __launch_bounds__() 80 …Points(const CeedInt num_elem, const CeedScalar *c_B, const CeedInt *points_per_elem, const CeedSc… in __launch_bounds__() 124 …ReadPoint<BASIS_NUM_COMP, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_elem * BASIS… in __launch_bounds__() 150 …Points(const CeedInt num_elem, const CeedScalar *c_B, const CeedInt *points_per_elem, const CeedSc… in __launch_bounds__() 183 …ReadPoint<BASIS_NUM_COMP, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_elem * BASIS… in __launch_bounds__() 212 …void GradAtPoints(const CeedInt num_elem, const CeedScalar *c_B, const CeedInt *points_per_elem, c… in __launch_bounds__() 268 …Points(const CeedInt num_elem, const CeedScalar *c_B, const CeedInt *points_per_elem, const CeedSc… in __launch_bounds__() 312 …ReadPoint<BASIS_NUM_COMP * BASIS_DIM, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_… in __launch_bounds__() 339 …Points(const CeedInt num_elem, const CeedScalar *c_B, const CeedInt *points_per_elem, const CeedSc… in __launch_bounds__() 372 …ReadPoint<BASIS_NUM_COMP * BASIS_DIM, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_… in __launch_bounds__()
|
| H A D | hip-ref-basis-tensor-at-points.h | 44 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in InterpAtPoints() argument 127 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in InterpTransposeAtPoints() argument 165 if (p >= points_per_elem[elem]) continue; in InterpTransposeAtPoints() 223 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in GradAtPoints() argument 312 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in GradTransposeAtPoints() argument 351 if (p >= points_per_elem[elem]) continue; in GradTransposeAtPoints()
|
| /libCEED/include/ceed/jit-source/cuda/ |
| H A D | cuda-ref-restriction-at-points.h | 16 …AtPointsTranspose(const CeedInt *__restrict__ indices, const CeedInt *__restrict__ points_per_elem, in AtPointsTranspose() argument 23 if (loc_node >= points_per_elem[elem]) continue; in AtPointsTranspose() 31 … const CeedInt *__restrict__ points_per_elem, const CeedInt *__restrict__ t_offsets, in AtPointsTranspose() argument 47 if (loc_node >= points_per_elem[elem]) continue; in AtPointsTranspose()
|
| H A D | cuda-shared-basis-tensor-at-points.h | 23 … CeedInt num_elem, const CeedScalar *__restrict__ c_B, const CeedInt *__restrict__ points_per_elem, in InterpAtPoints() argument 79 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ d_X, in InterpTransposeAtPoints() argument 123 …ReadPoint<BASIS_NUM_COMP, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_elem * BASIS… in InterpTransposeAtPoints() 149 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ d_X, in InterpTransposeAddAtPoints() argument 182 …ReadPoint<BASIS_NUM_COMP, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_elem * BASIS… in InterpTransposeAddAtPoints() 210 … CeedInt num_elem, const CeedScalar *__restrict__ c_B, const CeedInt *__restrict__ points_per_elem, in GradAtPoints() argument 266 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ d_X, in GradTransposeAtPoints() argument 310 …ReadPoint<BASIS_NUM_COMP * BASIS_DIM, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_… in GradTransposeAtPoints() 337 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ d_X, in GradTransposeAddAtPoints() argument 370 …ReadPoint<BASIS_NUM_COMP * BASIS_DIM, BASIS_NUM_PTS>(data, elem, i, points_per_elem[elem], 1, num_… in GradTransposeAddAtPoints()
|
| H A D | cuda-ref-basis-tensor-at-points.h | 44 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in InterpAtPoints() argument 127 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in InterpTransposeAtPoints() argument 165 if (p >= points_per_elem[elem]) continue; in InterpTransposeAtPoints() 223 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in GradAtPoints() argument 312 … const CeedInt *__restrict__ points_per_elem, const CeedScalar *__restrict__ coords, in GradTransposeAtPoints() argument 351 if (p >= points_per_elem[elem]) continue; in GradTransposeAtPoints()
|
| /libCEED/backends/cuda-gen/ |
| H A D | ceed-cuda-gen-operator.c | 179 CeedInt *points_per_elem; in CeedOperatorApplyAddCore_Cuda_gen() local 185 CeedCallBackend(CeedCalloc(num_elem, &points_per_elem)); in CeedOperatorApplyAddCore_Cuda_gen() 190 points_per_elem[e] = num_points_elem; in CeedOperatorApplyAddCore_Cuda_gen() 194 …CeedCallCuda(ceed, cudaMemcpy((void *)data->points.num_per_elem, points_per_elem, num_bytes, cudaM… in CeedOperatorApplyAddCore_Cuda_gen() 196 CeedCallBackend(CeedFree(&points_per_elem)); in CeedOperatorApplyAddCore_Cuda_gen() 607 CeedInt *points_per_elem; in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen() local 613 CeedCallBackend(CeedCalloc(num_elem, &points_per_elem)); in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen() 618 points_per_elem[e] = num_points_elem; in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen() 622 …CeedCallCuda(ceed, cudaMemcpy((void *)data->points.num_per_elem, points_per_elem, num_bytes, cudaM… in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen() 624 CeedCallBackend(CeedFree(&points_per_elem)); in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen() [all …]
|
| /libCEED/backends/hip-gen/ |
| H A D | ceed-hip-gen-operator.c | 124 CeedInt *points_per_elem; in CeedOperatorApplyAddCore_Hip_gen() local 130 CeedCallBackend(CeedCalloc(num_elem, &points_per_elem)); in CeedOperatorApplyAddCore_Hip_gen() 135 points_per_elem[e] = num_points_elem; in CeedOperatorApplyAddCore_Hip_gen() 139 …CeedCallHip(ceed, hipMemcpy((void *)data->points.num_per_elem, points_per_elem, num_bytes, hipMemc… in CeedOperatorApplyAddCore_Hip_gen() 141 CeedCallBackend(CeedFree(&points_per_elem)); in CeedOperatorApplyAddCore_Hip_gen() 586 CeedInt *points_per_elem; in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() local 592 CeedCallBackend(CeedCalloc(num_elem, &points_per_elem)); in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() 597 points_per_elem[e] = num_points_elem; in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() 601 …CeedCallHip(ceed, hipMemcpy((void *)data->points.num_per_elem, points_per_elem, num_bytes, hipMemc… in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() 603 CeedCallBackend(CeedFree(&points_per_elem)); in CeedOperatorLinearAssembleAddDiagonalAtPoints_Hip_gen() [all …]
|
| /libCEED/backends/cuda-ref/ |
| H A D | ceed-cuda-ref-restriction.c | 546 CeedInt max_points = elem_size, *offsets_padded, *points_per_elem; in CeedElemRestrictionCreate_Cuda() local 550 CeedCallBackend(CeedMalloc(num_elem, &points_per_elem)); in CeedElemRestrictionCreate_Cuda() 555 points_per_elem[i] = num_points; in CeedElemRestrictionCreate_Cuda() 580 …CeedCallBackend(CeedSetHostCeedIntArray(points_per_elem, CEED_OWN_POINTER, num_elem, &impl->h_poin… in CeedElemRestrictionCreate_Cuda()
|
| /libCEED/backends/hip-ref/ |
| H A D | ceed-hip-ref-restriction.c | 547 CeedInt max_points = elem_size, *offsets_padded, *points_per_elem; in CeedElemRestrictionCreate_Hip() local 551 CeedCallBackend(CeedMalloc(num_elem, &points_per_elem)); in CeedElemRestrictionCreate_Hip() 556 points_per_elem[i] = num_points; in CeedElemRestrictionCreate_Hip() 581 …CeedCallBackend(CeedSetHostCeedIntArray(points_per_elem, CEED_OWN_POINTER, num_elem, &impl->h_poin… in CeedElemRestrictionCreate_Hip()
|