Home
last modified time | relevance | path

Searched refs:atomicAdd (Results 1 – 15 of 15) sorted by relevance

/libCEED/include/ceed/jit-source/cuda/
H A Dcuda-atomic-add-fallback.h15 __device__ CeedScalar atomicAdd(CeedScalar *address, CeedScalar val) { in atomicAdd() function
H A Dcuda-ref-restriction-at-points.h25 …atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], u[loc_node + comp * RSTR_ELEM_SIZE * RSTR_NUM_ELEM + … in AtPointsTranspose()
H A Dcuda-gen-templates.h107 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[comp… in WriteLVecStandard1d()
121 atomicAdd(&d_v[ind + COMP_STRIDE * target_comp], r_v[target_comp]); in WriteLVecStandard1d_Single()
229 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[comp… in WriteLVecStandard2d()
245 atomicAdd(&d_v[ind + COMP_STRIDE * target_comp], r_v[target_comp]); in WriteLVecStandard2d_Single()
393 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[z + … in WriteLVecStandard3d()
411 atomicAdd(&d_v[ind + COMP_STRIDE * target_comp], r_v[target_node_z + target_comp * P_1D]); in WriteLVecStandard3d_Single()
H A Dcuda-ref-restriction-offset.h38 …atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], u[loc_node + comp * RSTR_ELEM_SIZE * RSTR_NUM_ELEM + … in OffsetTranspose()
H A Dcuda-ref-restriction-oriented.h42 atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], in OrientedTranspose()
H A Dcuda-ref-restriction-curl-oriented.h82 atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], value); in CurlOrientedTranspose()
140 atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], value); in CurlOrientedUnsignedTranspose()
H A Dcuda-ref-basis-tensor-at-points.h181 …for (CeedInt j = 0; j < Q; j++) atomicAdd(&out[(a * Q + (j + p) % Q) * post + c], chebyshev_x[(j +… in InterpTransposeAtPoints()
371 …for (CeedInt j = 0; j < Q; j++) atomicAdd(&out[(a * Q + (j + p) % Q) * post + c], chebyshev_x[(j +… in GradTransposeAtPoints()
/libCEED/include/ceed/jit-source/hip/
H A Dhip-shared-basis-tensor-at-points-templates.h78 …atomicAdd(&data.slice[comp * Q_1D + (i + data.t_id_x) % Q_1D], chebyshev_x[(i + data.t_id_x) % Q_1… in InterpTransposeAtPoints1d()
125 …atomicAdd(&data.slice[comp * Q_1D + (i + data.t_id_x) % Q_1D], chebyshev_x[(i + data.t_id_x) % Q_1… in GradTransposeAtPoints1d()
198 …if (data.t_id_x < Q_1D && data.t_id_y < Q_1D) atomicAdd(&data.slice[jj + ii * Q_1D], chebyshev_x[j… in InterpTransposeAtPoints2d()
274 …if (data.t_id_x < Q_1D && data.t_id_y < Q_1D) atomicAdd(&data.slice[jj + ii * Q_1D], chebyshev_x[j… in GradTransposeAtPoints2d()
359 …if (data.t_id_x < Q_1D && data.t_id_y < Q_1D) atomicAdd(&data.slice[jj + ii * Q_1D], chebyshev_x[j… in InterpTransposeAtPoints3d()
458 …if (data.t_id_x < Q_1D && data.t_id_y < Q_1D) atomicAdd(&data.slice[jj + ii * Q_1D], chebyshev_x[j… in GradTransposeAtPoints3d()
H A Dhip-ref-restriction-at-points.h25 …atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], u[loc_node + comp * RSTR_ELEM_SIZE * RSTR_NUM_ELEM + … in AtPointsTranspose()
H A Dhip-gen-templates.h106 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[comp… in WriteLVecStandard1d()
120 atomicAdd(&d_v[ind + COMP_STRIDE * target_comp], r_v[target_comp]); in WriteLVecStandard1d_Single()
227 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[comp… in WriteLVecStandard2d()
243 atomicAdd(&d_v[ind + COMP_STRIDE * target_comp], r_v[target_comp]); in WriteLVecStandard2d_Single()
390 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[z + … in WriteLVecStandard3d()
408 atomicAdd(&d_v[ind + COMP_STRIDE * target_comp], r_v[target_node_z + target_comp * P_1D]); in WriteLVecStandard3d_Single()
H A Dhip-ref-restriction-offset.h38 …atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], u[loc_node + comp * RSTR_ELEM_SIZE * RSTR_NUM_ELEM + … in OffsetTranspose()
H A Dhip-ref-restriction-oriented.h42 atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], in OrientedTranspose()
H A Dhip-ref-restriction-curl-oriented.h82 atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], value); in CurlOrientedTranspose()
140 atomicAdd(&v[ind + comp * RSTR_COMP_STRIDE], value); in CurlOrientedUnsignedTranspose()
H A Dhip-ref-basis-tensor-at-points.h181 …for (CeedInt j = 0; j < Q; j++) atomicAdd(&out[(a * Q + (j + p) % Q) * post + c], chebyshev_x[(j +… in InterpTransposeAtPoints()
371 …for (CeedInt j = 0; j < Q; j++) atomicAdd(&out[(a * Q + (j + p) % Q) * post + c], chebyshev_x[(j +… in GradTransposeAtPoints()
/libCEED/
H A DREADME.md227 However, some libCEED backends use non-deterministic operations, such as `atomicAdd` for increased …