code wiki / (root) / nx_sparse_tensor_test.nx

nx_sparse_tensor_test.nx

buildroot/runtime/nx_sparse_tensor_test.nx

6399 B158 linesdepth 4pulls 7 transitivereach 0 importersview sourcekind gate/proof
docsdependenciesstructsconstsfunctions

about

nx_sparse_tensor_test.nx -- algo-led correctness: sparse matmul must match dense matmul bit-exact across the same inputs. Oracle- verified per the cardinal: every algorithmic variant ships through nx_numeric_oracle vs the dense reference.

dependencies 6 imports · 0 importers

nx_syscalls.nx nx_tier.nx nx_tensor.nx nx_blas_i64.nx nx_sparse_tensor.nx nx_numeric_oracle.nx nx_sparse_tensor_test.nx

imports: nx_syscalls.nxnx_tier.nxnx_tensor.nxnx_blas_i64.nxnx_sparse_tensor.nxnx_numeric_oracle.nx

imported by: nobody (leaf or entry point)

call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown

main sys_mmap nx_t_alloc nx_dt_is_valid nx_dt_element_bytes sys_mmap ↻ nx_t_compute_strides_rowma nx_t_fill_zero nx_t_set_i64 nx_t_flat_offset nx_sp_alloc sys_mmap ↻ nx_sp_dense_to_csr nx_sp_csr_to_dense nx_t_fill_zero ↻ nx_no_check_bit_exact_i64 nx_no_witness_clear _no_structural_check nx_t_is_contiguous nx_blas_matmul nx_blas_gemm _blas_check_matmul nx_t_is_contiguous ↻ nx_sp_csr_matmul_dense nx_t_fill_zero ↻ nx_t_get_i64 nx_t_flat_offset ↻ nx_sp_sparsity_ratio_q10 nx_sp_estimated_speedup_q1 nx_sp_fmt_is_valid nx_sp_verdict_is_valid

structs

none

consts

none

functions

13func main() -> nx_int
calls 15: sys_mmapnx_t_allocnx_t_fill_zeronx_t_set_i64nx_sp_allocnx_sp_dense_to_csr+9