code wiki / _hdl_build / nx_simd_test.nx

nx_simd_test.nx

buildroot/runtime/_hdl_build/nx_simd_test.nx

2823 B60 linesdepth 3pulls 3 transitivereach 0 importersview sourcekind gate/prooftopic simd
docsdependenciesstructsconstsfunctions

about

nx_simd_test.nx -- prove the vectorizing transform is 1:1 with the scalar kernel and measure the op reduction (the SIMD win). The vector loop must produce the IDENTICAL output array to the scalar loop over a real, tail-bearing size, and issue ~N/VW vector ops vs the scalar 2N. Known answer: arrays identical AND vector ops << scalar ops -> exit 0.

dependencies 1 imports · 0 importers

nx_simd.nx nx_simd_test.nx

imports: nx_simd.nx

imported by: nobody (leaf or entry point)

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

main si_puts k_scalar k_vector sys_mmap v_load v_mul v_add v_store si_num

structs

none

consts

none

functions

9func si_puts(s: *u8) -> i64 { var n: i64 = 0; while s[n] != (0 as u8) { n = n + 1 } sys_write(1, s, n); return 0 }
called by 1: main
10func si_num(v: i64) -> i64
called by 1: main
19func main() -> i64