| /libCEED/backends/sycl-shared/ |
| H A D | ceed-sycl-shared-basis.sycl.cpp | 62 sycl::nd_range<3> kernel_range(global_range, local_range); in CeedBasisApplyTensor_Sycl_shared() local
72 cgh.parallel_for(kernel_range, *interp_kernel); in CeedBasisApplyTensor_Sycl_shared()
83 sycl::nd_range<3> kernel_range(global_range, local_range); in CeedBasisApplyTensor_Sycl_shared() local
95 cgh.parallel_for(kernel_range, *grad_kernel); in CeedBasisApplyTensor_Sycl_shared()
105 sycl::nd_range<3> kernel_range(global_range, local_range); in CeedBasisApplyTensor_Sycl_shared() local
115 cgh.parallel_for(kernel_range, *(impl->weight_kernel)); in CeedBasisApplyTensor_Sycl_shared()
|
| /libCEED/backends/sycl-ref/ |
| H A D | ceed-sycl-restriction.sycl.cpp | 35 sycl::range<1> kernel_range(num_elem * elem_size); in CeedElemRestrictionStridedNoTranspose_Sycl() local
40 sycl_queue.parallel_for<CeedElemRestrSyclStridedNT>(kernel_range, e, [=](sycl::id<1> node) { in CeedElemRestrictionStridedNoTranspose_Sycl()
62 sycl::range<1> kernel_range(num_elem * elem_size); in CeedElemRestrictionOffsetNoTranspose_Sycl() local
67 sycl_queue.parallel_for<CeedElemRestrSyclOffsetNT>(kernel_range, e, [=](sycl::id<1> node) { in CeedElemRestrictionOffsetNoTranspose_Sycl()
91 sycl::range<1> kernel_range(num_elem * elem_size); in CeedElemRestrictionStridedTranspose_Sycl() local
96 sycl_queue.parallel_for<CeedElemRestrSyclStridedT>(kernel_range, e, [=](sycl::id<1> node) { in CeedElemRestrictionStridedTranspose_Sycl()
121 sycl::range<1> kernel_range(num_nodes * num_comp); in CeedElemRestrictionOffsetTranspose_Sycl() local
126 sycl_queue.parallel_for<CeedElemRestrSyclOffsetT>(kernel_range, e, [=](sycl::id<1> id) { in CeedElemRestrictionOffsetTranspose_Sycl()
|
| H A D | ceed-sycl-ref-basis.sycl.cpp | 50 sycl::nd_range<1> kernel_range(global_range, local_range); in CeedBasisApplyInterp_Sycl() local
62 …cgh.parallel_for<CeedBasisSyclInterp<is_transpose>>(kernel_range, [=](sycl::nd_item<1> work_item, … in CeedBasisApplyInterp_Sycl()
153 sycl::nd_range<1> kernel_range(global_range, local_range); in CeedBasisApplyGrad_Sycl() local
165 …cgh.parallel_for<CeedBasisSyclGrad<is_transpose>>(kernel_range, [=](sycl::nd_item<1> work_item, sy… in CeedBasisApplyGrad_Sycl()
253 sycl::range<3> kernel_range(num_elem * num_quad_z, num_quad_y, num_quad_x); in CeedBasisApplyWeight_Sycl() local
259 sycl_queue.parallel_for<CeedBasisSyclWeight>(kernel_range, e, [=](sycl::item<3> work_item) { in CeedBasisApplyWeight_Sycl()
353 sycl::range<2> kernel_range(num_elem, v_size); in CeedBasisApplyNonTensorInterp_Sycl() local
359 sycl_queue.parallel_for<CeedBasisSyclInterpNT>(kernel_range, e, [=](sycl::id<2> indx) { in CeedBasisApplyNonTensorInterp_Sycl()
399 sycl::range<2> kernel_range(num_elem, v_size); in CeedBasisApplyNonTensorGrad_Sycl() local
405 sycl_queue.parallel_for<CeedBasisSyclGradNT>(kernel_range, e, [=](sycl::id<2> indx) { in CeedBasisApplyNonTensorGrad_Sycl()
[all …]
|
| H A D | ceed-sycl-ref-qfunction.sycl.cpp | 88 sycl::nd_range<1> kernel_range(rounded_Q, wg_size); in CeedQFunctionApply_Sycl() local
89 cgh.parallel_for(kernel_range, *(impl->QFunction)); in CeedQFunctionApply_Sycl()
|
| H A D | ceed-sycl-ref-operator.sycl.cpp | 847 sycl::range<1> kernel_range(num_elem * num_nodes); in CeedOperatorLinearDiagonal_Sycl() local
853 sycl_queue.parallel_for<CeedOperatorSyclLinearDiagonal>(kernel_range, e, [=](sycl::id<1> idx) { in CeedOperatorLinearDiagonal_Sycl()
1219 sycl::range<3> kernel_range(num_elem, block_size_y, block_size_x); in CeedOperatorLinearAssemble_Sycl() local
1224 sycl_queue.parallel_for<CeedOperatorSyclLinearAssemble>(kernel_range, e, [=](sycl::id<3> idx) { in CeedOperatorLinearAssemble_Sycl()
|
| /libCEED/backends/sycl/ |
| H A D | ceed-sycl-compile.sycl.cpp | 173 sycl::nd_range<3> kernel_range(global_range, local_range); in CeedRunKernelDimSharedSycl() local
185 cgh.parallel_for(kernel_range, *kernel); in CeedRunKernelDimSharedSycl()
|
| /libCEED/backends/sycl-gen/ |
| H A D | ceed-sycl-gen-operator.sycl.cpp | 142 sycl::nd_range<3> kernel_range(global_range, local_range); in CeedOperatorApplyAdd_Sycl_gen() local
152 cgh.parallel_for(kernel_range, *(impl->op)); in CeedOperatorApplyAdd_Sycl_gen()
|