ww/selfhost/test/smoke.ww

// selfhost/test/smoke.ww — end-to-end smoke for the selfhost path.
//
// Exercises the patterns the real ww-side compiler port will use:
//   - bump arena allocator (mem.ww shape)
//   - error idiom (T | str)
//   - struct of fn pointers + ctx pointer (the io.stream-style
//     polymorphism we use instead of interfaces)
//   - byte-level scanning that mirrors the hot path inside lex.ww
//   - strconv round-trip via the real stdlib
//
// `main` returns 42 when every check passes, 1..N on failure
// indicating which probe broke. The 990_selfhost test asserts 42.
//
// Note: only stack-local mutable state. Top-level `let` mutation
// requires a writable .data segment in w6l, which is a separate
// task; until then we exercise polymorphism via ctx pointers, which
// is what the real port wants anyway.

package test;

import os;
import strconv;
import ascii;

// --- bump arena ---------------------------------------------------------

type arena = struct {
	buf: *u8,
	off: u64,
	cap: u64,
};

// In-place init. Returning a 24-byte struct by value isn't yet
// supported in w6c (SysV requires a hidden return-slot pointer for
// structs >16 bytes), so we initialize through a pointer like the
// real compiler does today.
fn arena_init(a: *arena, buf: *u8, cap: u64) void = {
	a.buf = buf;
	a.off = 0u64;
	a.cap = cap;
};

fn arena_alloc(a: *arena, n: u64) *u8 = {
	if (n > a.cap - a.off) { return nil; };
	let p: *u8 = a.buf + a.off;
	a.off += n;
	return p;
};

// --- (i32 | str) error idiom -------------------------------------------

fn checked_div(num: i32, den: i32) (i32 | str) = {
	if (den == 0) { return "div by zero"; };
	return num / den;
};

// --- struct-of-fn-pointer polymorphism ---------------------------------
//
// A trivial "writer" abstraction: a function pointer plus a context.
// This mirrors how io.stream / Plan 9 Bio work. The ctx pointer lets
// the implementation own its own state without a global.

type counter = struct {
	n: i32,
};

type writer = struct {
	ctx:  *void,
	emit: fn(ctx: *void, b: u8) void,
};

fn count_emit(ctx: *void, b: u8) void = {
	let c: *counter = ctx: *counter;
	c.n += 1;
};

// --- size/align/offset typed-builtin fixtures (#42) --------------------

type point = struct {
	x: i32,
	y: i32,
};

// Mixed-alignment struct: i8 lays at 0, then i64 needs to skip to
// offset 8 (the i64's natural align). Probe asserts both ends.
type mixalign = struct {
	tag: i8,
	val: i64,
};

// --- byte scanner like lex.ww's hot path -------------------------------

fn count_digits(s: str) i32 = {
	let i: i32 = 0;
	let n: i32 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		if (c >= 48u8) {
			if (c <= 57u8) { n += 1; };
		};
		i += 1;
	};
	return n;
};

// --- entry --------------------------------------------------------------

export fn main() i32 = {
	// Probe 1 — arena hands out distinct pointers, refuses oversize.
	let buf: [256]u8;
	let a: arena;
	arena_init(&a, buf.ptr, 256u64);
	let p1: *u8 = arena_alloc(&a, 32u64);
	let p2: *u8 = arena_alloc(&a, 32u64);
	if (p1 == nil) { return 1; };
	if (p2 == nil) { return 2; };
	if (p1 == p2) { return 3; };
	let p3: *u8 = arena_alloc(&a, 1024u64);
	if (p3 != nil) { return 4; };

	// Probe 2 — error union both ways.
	let r_ok: (i32 | str) = checked_div(84, 2);
	let r_bad: (i32 | str) = checked_div(1, 0);
	let acc: i32 = 0;
	match (r_ok) {
		case let v: i32 => acc = v;
		case let e: str => return 5;
	};
	if (acc != 42) { return 6; };
	match (r_bad) {
		case let v: i32 => return 7;
		case let e: str => acc = e.len: i32;
	};
	if (acc != 11) { return 8; };	// len("div by zero") == 11

	// Probe 3 — struct-of-fn-pointer dispatch via ctx pointer.
	let c: counter = counter { n = 0 };
	let w: writer = writer { ctx = (&c): *void, emit = count_emit };
	w.emit(w.ctx, 65u8);
	w.emit(w.ctx, 66u8);
	w.emit(w.ctx, 67u8);
	if (c.n != 3) { return 9; };

	// Probe 4 — byte scan over a literal.
	let dn: i32 = count_digits("ww123abc");
	if (dn != 3) { return 10; };

	// Probe 5 — strconv round-trip via the real stdlib.
	let s: str = strconv.i64tos(4242i64, strconv.base.DEC);
	if (s.len != 4) { return 11; };
	if (s.ptr[0] != 52u8) { return 12; };	// '4'
	if (s.ptr[3] != 50u8) { return 13; };	// '2'

	// Probe 6 — ascii classifications (rune-taking, Hare-shaped).
	if (!ascii.isdigit(53)) { return 14; };		// '5'
	if (ascii.isdigit(65))  { return 15; };		// 'A' is not a digit
	if (!ascii.isalpha(122)) { return 16; };	// 'z'
	if (!ascii.isxdigit(70)) { return 17; };	// 'F'
	if (ascii.isxdigit(71))  { return 18; };	// 'G' is not hex
	if (ascii.tolower(65) != 97) { return 19; };	// 'A' -> 'a'
	if (ascii.toupper(122) != 90) { return 20; };	// 'z' -> 'Z'

	// Probe 7 — file open/read via the new os APIs. /proc/self/cmdline
	// always exists on Linux, no write side, and is non-empty.
	let path: str = "/proc/self/cmdline";
	// Use raw os.open here (returns i32 with -errno) for the same
	// reason as os.read below: probe 6 in 990_selfhost compiles
	// smoke.ww standalone (no `use` expansion), so cross-module type
	// references like `os.oserror` and `os.flag` don't resolve at
	// that step. RDONLY is 0; passing the literal keeps the call
	// site standalone-compilable to byte-identical asm on both
	// compilers.
	let fd: i32 = os.open(path, 0, 0i32);
	if (fd < 0) { return 21; };
	let rbuf: [128]u8;
	// Use raw os.read here (single syscall, plain i64) instead of
	// os.readall: the 990 cgen-match probe compiles smoke.ww
	// standalone without `use os;` expansion, so cross-module type
	// references like `os.oserror` can't be resolved.
	let n: i64 = os.read(fd, rbuf.ptr, 128u64);
	os.close(fd);
	if (n <= 0i64) { return 22; };

	// Probe 8 — size(T) / align(T) / offset(e.f) typed-builtin folds
	// (#42). Each call folds to an N_INTLIT at check time; cgen
	// materialises the literal as a plain `MOVQ $N, AX`. Mirrors
	// cstage cmd/wcc/check.c:907-960 byte-for-byte on this corpus.
	if (size(str) != 16) { return 23; };
	if (size(i64) != 8)  { return 24; };
	if (size(i32) != 4)  { return 25; };
	if (align(i64) != 8) { return 26; };
	if (align(i32) != 4) { return 27; };
	// Initialize struct locals explicitly so the cgen path doesn't
	// drift from cstage on bare `let X: T;` zero-init (pre-existing
	// wwstage divergence outside #42).
	let pt: point = point { x = 0, y = 0 };
	if (offset(pt.x) != 0) { return 28; };
	if (offset(pt.y) != 4) { return 29; };
	let mx: mixalign = mixalign { tag = 0i8, val = 0i64 };
	if (offset(mx.tag) != 0) { return 30; };
	if (offset(mx.val) != 8) { return 31; };	// align-padded to 8
	// Width breadth: smallest prim, ptr, slice, struct (8B + padded),
	// covering astsize's TPTR/TSLICE/TNAME-resolve-to-struct arms.
	if (size(i8) != 1)        { return 32; };
	if (align(i8) != 1)       { return 33; };
	if (size(*i32) != 8)      { return 34; };
	if (size([]i32) != 24)    { return 35; };
	if (size(point) != 8)     { return 36; };
	if (size(mixalign) != 16) { return 37; };

	return 42;
};