code wiki / _hdl_build / nx_simd_isa_test.nx

nx_simd_isa_test.nx

buildroot/runtime/_hdl_build/nx_simd_isa_test.nx

3531 B73 linesdepth 3pulls 3 transitivereach 0 importersview sourcekind gate/prooftopic simd
docsdependenciesstructsconstsfunctions

about

nx_simd_isa_test.nx -- the vectorized FMA kernel as real vector MACHINE CODE, proven by EXECUTION on the sovereign SIMD emulator. Memory image holds a|b|d|c (N each). For each block the vectorizer EMITS 6 vector instructions (3 VLOAD, VMUL, VADD, VSTORE) and sv_run executes them; the resulting c region must equal the scalar c, 1:1, including the scalar tail. Known answer: arrays identical AND vector instructions << scalar ops -> exit 0.

dependencies 1 imports · 0 importers

nx_simd_isa.nx nx_simd_isa_test.nx

imports: nx_simd_isa.nx

imported by: nobody (leaf or entry point)

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

main sx_puts sv_emit sv_run sx_num

structs

none

consts

none

functions

10func sx_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
11func sx_num(v: i64) -> i64
called by 1: main
20func main() -> i64