nx_nofloat_gemm.nx
buildroot/runtime/nx_nofloat_gemm.nx
about
nx_nofloat_gemm.nx -- canonical sovereign INTEGER SIMD GEMM (vpmaddwd / __i16x16_madd):
EXACT + DETERMINISTIC + fast = the no-float compute lever every Nishi model rides. Pure funcs, no main.
Extracted from nx_nofloat_gemm_gate for debt-free reuse (gate + nx_gpu_bench both import this).
__i16x16_madd accumulates 16 signed-int16 products into an int32x8 vector (vpmaddwd + vpaddd, proven VEX);
integer add is associative -> bit-exact + deterministic (the no-float exceed axis, unlike float cuBLAS).
license_tier: ORIGINAL
dependencies 1 imports · 3 importers
imports: nx_syscalls.nx
imported by: nx_gpu_bench.nxnx_nofloat_gemm_gate.nxnx_nofloat_simd_dot_probe_gate.nx
structs
| none |
consts
| none |
functions
| 9 | func pack2(buf: *u8, idx: i64, val: i64) -> i64 { buf[idx*2]=(val) as u8; buf[idx*2+1]=(val>>8) as u8; return 0 } |
| 15 | func i32x8_hsum(acc: *u8) -> i64 { let p: *i32 = acc as *i32; var s: i64=0; var i: i64=0; while i<8 { s = s + (p[i] as i64); i=i+1 } return s } called by 1: simd_imm |
| 17 | func scalar_imm(ai: *i64, bi: *i64, c: *i64, M: i64, N: i64, K: i64) -> i64 |
| 26 | func simd_imm(a16: *u8, b16: *u8, c: *i64, acc: *u8, M: i64, N: i64, K: i64) -> i64 |