ww/test/wcc/991_w6a_ww.c

/*
 * Explicit bootstrap gate: the C-built `w6a` and the ww-built `w6a_ww`
 * must produce byte-identical ELF .o output over the corpus. Any
 * divergence is a port bug in selfhost/cmd/w6a/. The corpus pairs
 * hand-written rt/X.s (small, varied operands) with cgen-generated .s
 * (large — exercises every encoding path w6c emits).
 */
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>
#include <glob.h>
#include <sys/wait.h>

static int
runwait(const char *cmd)
{
	int rc = system(cmd);
	if (rc == -1) return -1;
	if (WIFEXITED(rc)) return WEXITSTATUS(rc);
	return 1;
}

static const char *
absbin(void)
{
	const char *b = getenv("BIN");
	if (!b) b = "out/bin";
	if (b[0] == '/') return b;
	static char buf[2048];
	char cwd[1024];
	if (getcwd(cwd, sizeof cwd) == NULL) return NULL;
	snprintf(buf, sizeof buf, "%s/%s", cwd, b);
	return buf;
}

static int
slurp(const char *path, char **outbuf, size_t *outlen)
{
	FILE *f = fopen(path, "rb");
	if (!f) return -1;
	fseek(f, 0, SEEK_END);
	long n = ftell(f);
	fseek(f, 0, SEEK_SET);
	if (n < 0) { fclose(f); return -1; }
	char *b = malloc((size_t)n + 1);
	if (!b) { fclose(f); return -1; }
	if (fread(b, 1, (size_t)n, f) != (size_t)n) { free(b); fclose(f); return -1; }
	b[n] = '\0';
	fclose(f);
	*outbuf = b;
	*outlen = (size_t)n;
	return 0;
}

static int
diff_one(const char *bin, const char *src)
{
	char co[64], wo[64], cmd[2048];
	char *bc = NULL, *bw = NULL;
	size_t nc = 0, nw = 0;
	int rc = -1;
	snprintf(co, sizeof co, "/tmp/wwa_%d_c.o", getpid());
	snprintf(wo, sizeof wo, "/tmp/wwa_%d_w.o", getpid());

	snprintf(cmd, sizeof cmd, "timeout 180 %s/w6a -o %s %s 2>/dev/null", bin, co, src);
	if (runwait(cmd) != 0) {
		fprintf(stderr, "w6a_ww FAIL: C w6a errored on %s\n", src);
		goto cleanup;
	}
	snprintf(cmd, sizeof cmd, "timeout 180 %s/w6a_ww -o %s %s 2>/dev/null", bin, wo, src);
	if (runwait(cmd) != 0) {
		fprintf(stderr, "w6a_ww FAIL: ww w6a errored on %s\n", src);
		goto cleanup;
	}
	if (slurp(co, &bc, &nc) < 0 || slurp(wo, &bw, &nw) < 0) {
		fprintf(stderr, "w6a_ww FAIL: cannot read .o for %s\n", src);
	} else if (nc == 0) {
		fprintf(stderr, "w6a_ww FAIL: empty .o for %s\n", src);
	} else if (nc != nw || memcmp(bc, bw, nc) != 0) {
		fprintf(stderr, "w6a_ww FAIL: %s — C %zu vs ww %zu bytes\n",
		    src, nc, nw);
	} else
		rc = 0;

cleanup:
	free(bc); free(bw);
	/* co/wo are reused across corpus files; a stale .o left behind
	 * would be slurped as a later file's output. */
	int cleanfail = 0;
	if (unlink(co) != 0 && errno != ENOENT) {
		perror(co);
		cleanfail = 1;
	}
	if (unlink(wo) != 0 && errno != ENOENT) {
		perror(wo);
		cleanfail = 1;
	}
	if (cleanfail && rc == 0)
		rc = -1;
	return rc;
}

int
main(void)
{
	const char *bin = absbin();
	if (!bin) return 1;

	char cwd[1024];
	if (getcwd(cwd, sizeof cwd) == NULL) return 1;

	/* #93 sep layout: the selfhost tools no longer leave a single
	 * next-to-source main.s. Self-build one real directory package and
	 * feed EVERY emitted
	 * <tool>.sepwork/<pkg>.s as
	 * the generated-asm corpus — the root tool code plus each dep
 * package's .s (richer than the old 3 monolithic main.s). */
	char stem[256], cmd[4096];
	int fail = 0;
	int n = 0;
	int gen = 0;
	snprintf(stem, sizeof stem, "/tmp/wwa_%d_sh", getpid());
	snprintf(cmd, sizeof cmd,
	    "timeout 400 %s/ww build -o %s "
	    "%s/selfhost/cmd/w6a >/dev/null 2>&1",
	    bin, stem, cwd);
	if (runwait(cmd) != 0) {
		fprintf(stderr,
		    "w6a_ww FAIL: cannot sep-build selfhost/cmd/w6a\n");
		/* a failed build can leave a partial stem/.sepwork tree. */
		fail = 1;
		goto cleanup;
	}

	/* hand-written runtime — small, covers MOVQ/LEAQ/CALL/SYSCALL/RET;
	 * exists in-tree regardless of layout (not a casualty). */
	const char *rtfiles[] = {
		"rt/start.s", "rt/syscall.s", "rt/abort.s",
		"rt/alloc.s", "rt/streq.s", NULL,
	};
	for (int i = 0; rtfiles[i]; i++) {
		char p[2048];
		snprintf(p, sizeof p, "%s/%s", cwd, rtfiles[i]);
		if (diff_one(bin, p) != 0) fail++;
		n++;
	}

	/* generated .s from the real tool's sep build — large, exercises
	 * every encoding w6c emits (the long tail of operand shapes). */
	glob_t g;
	char pat[320];
	snprintf(pat, sizeof pat, "%s.sepwork/*.s", stem);
	if (glob(pat, 0, NULL, &g) == 0) {
		for (size_t i = 0; i < g.gl_pathc; i++) {
			/* skip empty units (e.g. an import-only rt.s). */
			FILE *f = fopen(g.gl_pathv[i], "rb");
			long sz = -1;
			if (f) { fseek(f, 0, SEEK_END); sz = ftell(f); fclose(f); }
			if (sz <= 0) continue;
			if (diff_one(bin, g.gl_pathv[i]) != 0) fail++;
			n++;
			gen++;
		}
		globfree(&g);
	}
	/* the PASS banner claims generated-asm coverage; an empty glob
	 * (or all-empty units) must not go green on the rt corpus alone. */
	if (gen == 0) {
		fprintf(stderr,
		    "w6a_ww FAIL: no non-empty generated .s under %s.sepwork\n",
		    stem);
		fail++;
	}

cleanup:
	snprintf(cmd, sizeof cmd, "rm -rf %s.sepwork %s", stem, stem);
	int cleanfail = runwait(cmd) != 0;
	if (cleanfail)
		fprintf(stderr, "w6a_ww: cleanup of %s{,.sepwork} failed\n",
		    stem);

	if (fail) {
		fprintf(stderr, "w6a_ww: %d/%d diff(s) failed\n", fail, n);
		return 1;
	}
	if (cleanfail)
		return 1;
	printf("w6a_ww: byte-identical to C w6a on %d corpus files "
	    "(rt/*.s + selfhost/cmd/w6a sep-built .s)\n", n);
	return 0;
}