nx_i8dot32_kat.nx
buildroot/runtime/nx_i8dot32_kat.nx
about
nx_i8dot32_kat.nx -- KAT for the new __f32_i8dot32 SSE intrinsic.
__f32_i8dot32(a:*i8[32], b:*f32[32]) = sum_{j<32} sext(a[j]) * b[j].
Compared to a scalar hardware-op reference; exact-int regime so the
4-lane-parallel accumulation is bit-identical to sequential.
expect_exit: 0 (77=mismatch)
dependencies 3 imports · 0 importers
imports: nx_syscalls.nxnx_f32.nxnx_f32_cvt.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| none |
functions
| 10 | func k_st4(p: *u8, idx: i64, bits: i64) -> i64 called by 1: main |
| 18 | func main() -> i64 |