| /libCEED/backends/hip-shared/ |
| H A D | ceed-hip-shared-basis.c | 177 void *grad_args[] = {(void *)&num_elem, &data->d_interp_1d, &d_grad_1d, &d_u, &d_v}; in CeedBasisApplyTensorCore_Hip_shared() local
187 elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Hip_shared()
189 …DimShared_Hip(ceed, data->Grad, NULL, grid, thread_1d, 1, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Hip_shared()
199 elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Hip_shared()
201 …d_Hip(ceed, data->Grad, NULL, grid, thread_1d, thread_1d, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Hip_shared()
210 elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Hip_shared()
212 …d_Hip(ceed, data->Grad, NULL, grid, thread_1d, thread_1d, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Hip_shared()
436 …void *grad_args[] = {(void *)&num_elem, &data->d_chebyshev_interp_1d, &data->d_points_per_elem, … in CeedBasisApplyAtPointsCore_Hip_shared() local
446 … thread_1d, 1, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyAtPointsCore_Hip_shared()
448 …d_Hip(ceed, data->GradAtPoints, NULL, grid, thread_1d, 1, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyAtPointsCore_Hip_shared()
[all …]
|
| /libCEED/backends/cuda-shared/ |
| H A D | ceed-cuda-shared-basis.c | 112 void *grad_args[] = {(void *)&num_elem, &data->d_interp_1d, &d_grad_1d, &d_u, &d_v}; in CeedBasisApplyTensorCore_Cuda_shared() local
122 elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Cuda_shared()
124 …imShared_Cuda(ceed, data->Grad, NULL, grid, thread_1d, 1, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Cuda_shared()
135 … thread_1d, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Cuda_shared()
137 …_Cuda(ceed, data->Grad, NULL, grid, thread_1d, thread_1d, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Cuda_shared()
146 … thread_1d, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Cuda_shared()
148 …_Cuda(ceed, data->Grad, NULL, grid, thread_1d, thread_1d, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyTensorCore_Cuda_shared()
372 …void *grad_args[] = {(void *)&num_elem, &data->d_chebyshev_interp_1d, &data->d_points_per_elem, &d… in CeedBasisApplyAtPointsCore_Cuda_shared() local
382 … thread_1d, 1, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyAtPointsCore_Cuda_shared()
384 …_Cuda(ceed, data->GradAtPoints, NULL, grid, thread_1d, 1, elems_per_block, shared_mem, grad_args)); in CeedBasisApplyAtPointsCore_Cuda_shared()
[all …]
|
| /libCEED/backends/hip-ref/ |
| H A D | ceed-hip-ref-basis.c | 57 …void *grad_args[] = {(void *)&num_elem, (void *)&is_transpose, &data->d_interp_1d, &data->… in CeedBasisApplyCore_Hip() local
60 CeedCallBackend(CeedRunKernel_Hip(ceed, data->Grad, num_elem, block_size, grad_args)); in CeedBasisApplyCore_Hip()
215 …void *grad_args[] = {(void *)&num_elem, &data->d_chebyshev_interp_1d, &data->d_points_per_… in CeedBasisApplyAtPointsCore_Hip() local
218 …is_transpose ? data->GradTransposeAtPoints : data->GradAtPoints, num_elem, block_size, grad_args)); in CeedBasisApplyAtPointsCore_Hip()
294 void *grad_args[] = {(void *)&num_elem, &data->d_grad, &d_u, &d_v}; in CeedBasisApplyNonTensorCore_Hip() local
298 …edRunKernelDim_Hip(ceed, data->DerivTranspose, grid, block_size_x, 1, elems_per_block, grad_args)); in CeedBasisApplyNonTensorCore_Hip()
300 …ackend(CeedRunKernelDim_Hip(ceed, data->Deriv, grid, block_size_x, 1, elems_per_block, grad_args)); in CeedBasisApplyNonTensorCore_Hip()
|
| /libCEED/backends/cuda-ref/ |
| H A D | ceed-cuda-ref-basis.c | 57 …void *grad_args[] = {(void *)&num_elem, (void *)&is_transpose, &data->d_interp_1d, &data->… in CeedBasisApplyCore_Cuda() local
60 CeedCallBackend(CeedRunKernel_Cuda(ceed, data->Grad, num_elem, block_size, grad_args)); in CeedBasisApplyCore_Cuda()
216 …void *grad_args[] = {(void *)&num_elem, &data->d_chebyshev_interp_1d, &data->d_points_per_… in CeedBasisApplyAtPointsCore_Cuda() local
219 …is_transpose ? data->GradTransposeAtPoints : data->GradAtPoints, num_elem, block_size, grad_args)); in CeedBasisApplyAtPointsCore_Cuda()
295 void *grad_args[] = {(void *)&num_elem, &data->d_grad, &d_u, &d_v}; in CeedBasisApplyNonTensorCore_Cuda() local
299 …dRunKernelDim_Cuda(ceed, data->DerivTranspose, grid, block_size_x, 1, elems_per_block, grad_args)); in CeedBasisApplyNonTensorCore_Cuda()
301 …ckend(CeedRunKernelDim_Cuda(ceed, data->Deriv, grid, block_size_x, 1, elems_per_block, grad_args)); in CeedBasisApplyNonTensorCore_Cuda()
|