code wiki / _hdl_build / nx_rv64_fast_gate.nx
nx_rv64_fast_gate.nx
buildroot/runtime/_hdl_build/nx_rv64_fast_gate.nx
about
nx_rv64_fast_gate.nx -- GATE: prove the predecoded interpreter (nx_rv64_fast) is (a) EQUIVALENT to the golden
behavioral sim (rv64im_min_sim) on assembled programs, and (b) FASTER (measured speedup). Both run the SAME
sovereign-assembled binaries; each program stores its result to a scratch word the gate reads back from each
engine's memory. Equivalence = same result; then the big compute loop is timed on both -> measured MIPS + speedup.
This is the honest first perf-engineering rung: decode-once beats decode-per-step, validated against the golden model.
expect_exit:0
dependencies 11 imports · 0 importers
diagram shows first 10 each side; +1 more imports, +0 more importers in the complete lists below.
imports: nx_syscalls.nxnishi_hdl_primitives.nxrv64im_min_decoder.nxrv64im_min_alu.nxrv64im_min_regfile.nxrv64im_min_csr.nxrv64im_min_clint.nxrv64im_min_uart.nxrv64im_min_sim.nxnx_rv64_asm.nxnx_rv64_fast.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
| 23 | const GMEM_BASE: i64 = 0x80000000 |
| 24 | const GMEM_SIZE: i64 = 65536 |
| 25 | const GSCRATCH_OFF: i64 = 0x8000 // 0x80008000 - base |
functions
| 19 | func g_puts(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(1,s,n); return 0 } |
| 20 | func g_pn(v: i64) -> i64 { let b: *u8=sys_mmap(28); var x: i64=v; if x<0{b[0]=45;sys_write(1,b,1);x=0-x} if x==0{b[0]=48;sys_write(1,b,1);return 0} var d: i64=0; var y: i64=x; while y>0{d=d+1;y=y/10} var i: i64=d-1; y=x; while i>=0{b[i]=(48+(y%10)) as u8;y=y/10;i=i-1} sys_write(1,b,d); return 0 } |
| 21 | func ck(name: *u8, c: i64) -> i64 { if c==1 { g_puts(" PASS " as *u8) } else { g_puts(" FAIL " as *u8) } g_puts(name); g_puts("\n" as *u8); return c } |
| 28 | func run_on_sim(code: *u8, nb: i64, steps_out: *i64, us_out: *i64) -> i64 called by 1: main calls 9: sys_mmapnx_rv64im_rf_initnx_rv64im_csr_initnx_clint_initnx_uart_initnx_rv64im_sim_init+3 |
| 53 | func run_on_fast(code: *u8, nb: i64, steps_out: *i64, us_out: *i64) -> i64 |
| 67 | func asm_it(src: *u8, out: *u8) -> i64 { return rvasm_assemble_str(src, out, 65536) } |
| 68 | func mips_of(steps: i64, us: i64) -> i64 { if us<=0 { return 0 } return steps/us } // steps/us == millions/sec called by 1: main |
| 70 | func main() -> i64 |