w6c+wwstage: read array field of a global struct (#249 BUG B)

Reading an array-typed field of a module-global struct value (`G.arr[i]`)
silently miscompiled: the N_INDEX fallback's cg_dotbase_addr (cstage) /
dotbaseaddr (wwstage) helper — the #135 sibling that computes &(s.field)
for a `[N]T` field — had no module-global-struct base arm. cstage emitted
`LEAQ (BP)` (localfind returns 0 for a global, so it read the stack frame
→ 0); wwstage's localfindnode returned nil and the fallback keyed on the
FIELD name, so it returned false and the caller's cgexpr(N_DOT) loaded the
field VALUE as a pointer → SEGFAULT. The .data was already correct
(emit_struct_lit_bytes #129 A.3); only the READ base address was wrong.

Both stages now emit `LEAQ name(SB) (+ ADDQ field_off)` for a global
value-struct base, mirroring the scalar global-field read (cgen.c:7532);
const globals resolve via def_isstructdef. Symmetric both stages (rule
10), byte-identical .s. Unblocks base64's `const std_encoding.encmap[i]`
reads (#22).

Test 949_structlit_arrfield_run: global `let`/`def` struct array-field
read, cstage run + cs==ww byte-id.

Makefile

@@ -402,6 +402,7 @@ TESTS	= $(BIN)/test_smoke $(BIN)/test_lex $(BIN)/test_parse $(BIN)/test_check \
 	  $(BIN)/test_deref_narrow_run \
 	  $(BIN)/test_idx_compound_run \
 	  $(BIN)/test_dotbase_arr_run \
+	  $(BIN)/test_structlit_arrfield_run \
 	  $(BIN)/test_continue_run \
 	  $(BIN)/test_callret_unsigned_arith_run \
 	  $(BIN)/test_sar_shr_run \
@@ -1592,6 +1593,11 @@ $(BIN)/test_dotbase_arr_run: test/wcc/949_dotbase_arr_run.c $(BIN)/ww \
 		$(LIB)/libwwrt.a | $(BIN)
 	$(CC) $(CFLAGS) -o $@ $<
 
+$(BIN)/test_structlit_arrfield_run: test/wcc/949_structlit_arrfield_run.c \
+		$(BIN)/ww $(BIN)/w6c $(BIN)/w6c_ww $(BIN)/w6a $(BIN)/w6l \
+		$(LIB)/libwwrt.a | $(BIN)
+	$(CC) $(CFLAGS) -o $@ $<
+
 $(BIN)/test_continue_run: test/wcc/911_continue_run.c $(BIN)/ww \
 		$(BIN)/w6c $(BIN)/w6c_ww $(BIN)/w6a $(BIN)/w6l \
 		$(LIB)/libwwrt.a | $(BIN)

cmd/w6c/cgen.c

@@ -1754,10 +1754,20 @@ cg_dotbase_addr(Cg *c, Node *base, int dst_reg, Local *locals)
 	if (ft == NULL || ft->kind != TY_ARRAY) return 0;
 	int inner_off = localfind(locals, inner->str);
 	int foff = (int)f->offset;
+	/* #249 (sibling of #135): a module-GLOBAL struct value base. localfind
+	 * returns 0 for a global, so the BP-rel form below would emit `LEAQ
+	 * (BP)` (read the stack frame, not the global). Resolve the same way
+	 * the scalar N_DOT global-field read does (cgen.c:7532) — LEAQ
+	 * name(SB) + field offset. const globals are def_isstructdef. */
 	if (viaptr) {
 		ins2(c, A_MOVQ, amem(D_BP, inner_off), areg(dst_reg));
 		if (foff != 0)
 			ins2(c, A_ADDQ, aimm(foff), areg(dst_reg));
+	} else if (inner_off == 0 && (let_islet(inner->str)
+	    || def_isstructdef(inner->str))) {
+		ins2(c, A_LEAQ, masym(c, inner->str), areg(dst_reg));
+		if (foff != 0)
+			ins2(c, A_ADDQ, aimm(foff), areg(dst_reg));
 	} else {
 		ins2(c, A_LEAQ, amem(D_BP, inner_off + foff),
 		    areg(dst_reg));

selfhost/cmd/w6c/main.combined.ww

@@ -20025,6 +20025,7 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 	// 8 bytes as if it were a pointer-var — wrong shape (cstage
 	// sister fix in cg_dotbase_addr).
 	let lc: *local = localfindnode(c, inner.str);
+	let isglobal: bool = false;
 	if (lc == nil) {
 		let gt: *node = letvartnode(c, base.str);
 		if (gt != nil && gt.kind == nkind.N_TARRAY) {
@@ -20035,7 +20036,12 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 			emitline("\n");
 			return true;
 		};
-		return false;
+		// #249 (sibling of #135): inner is a module-GLOBAL struct value
+		// (let/def), not a local — lc is nil but inner.type_ is a valid
+		// struct. Resolve the field below and emit a global base (LEAQ
+		// name(SB)). A non-struct inner (e.g. an SK_USE module qualifier,
+		// type ty_err) falls through the struct gate to `return false`.
+		isglobal = true;
 	};
 	let bu: *tinfo = inner.type_: *tinfo;
 	for (bu != nil && bu.kind == tykind.TY_NAMED) { bu = bu.under; };
@@ -20072,7 +20078,8 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 	for (ft != nil && ft.kind == tykind.TY_NAMED) { ft = ft.under; };
 	if (ft == nil) { return false; };
 	if (ft.kind != tykind.TY_ARRAY) { return false; };
-	let innoff: i64 = lc.off: i64;
+	let innoff: i64 = 0;
+	if (lc != nil) { innoff = lc.off: i64; };
 	if (viaptr) {
 		emitline("\tMOVQ\t");
 		emitoff(innoff);
@@ -20086,6 +20093,21 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 			emitline(dstreg);
 			emitline("\n");
 		};
+	} else if (isglobal) {
+		// #249: LEAQ name(SB) + field offset. Mirror cstage
+		// cg_dotbase_addr's global value-struct arm.
+		emitline("\tLEAQ\t");
+		emitsymname(c, inner.str);
+		emitline("(SB), ");
+		emitline(dstreg);
+		emitline("\n");
+		if (foff != 0) {
+			emitline("\tADDQ\t$");
+			emitint(foff);
+			emitline(", ");
+			emitline(dstreg);
+			emitline("\n");
+		};
 	} else {
 		emitline("\tLEAQ\t");
 		emitoff(innoff + foff);

selfhost/cmd/wcc/cgenexpr.ww

@@ -918,6 +918,7 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 	// 8 bytes as if it were a pointer-var — wrong shape (cstage
 	// sister fix in cg_dotbase_addr).
 	let lc: *local = localfindnode(c, inner.str);
+	let isglobal: bool = false;
 	if (lc == nil) {
 		let gt: *node = letvartnode(c, base.str);
 		if (gt != nil && gt.kind == nkind.N_TARRAY) {
@@ -928,7 +929,12 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 			emitline("\n");
 			return true;
 		};
-		return false;
+		// #249 (sibling of #135): inner is a module-GLOBAL struct value
+		// (let/def), not a local — lc is nil but inner.type_ is a valid
+		// struct. Resolve the field below and emit a global base (LEAQ
+		// name(SB)). A non-struct inner (e.g. an SK_USE module qualifier,
+		// type ty_err) falls through the struct gate to `return false`.
+		isglobal = true;
 	};
 	let bu: *tinfo = inner.type_: *tinfo;
 	for (bu != nil && bu.kind == tykind.TY_NAMED) { bu = bu.under; };
@@ -965,7 +971,8 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 	for (ft != nil && ft.kind == tykind.TY_NAMED) { ft = ft.under; };
 	if (ft == nil) { return false; };
 	if (ft.kind != tykind.TY_ARRAY) { return false; };
-	let innoff: i64 = lc.off: i64;
+	let innoff: i64 = 0;
+	if (lc != nil) { innoff = lc.off: i64; };
 	if (viaptr) {
 		emitline("\tMOVQ\t");
 		emitoff(innoff);
@@ -979,6 +986,21 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 			emitline(dstreg);
 			emitline("\n");
 		};
+	} else if (isglobal) {
+		// #249: LEAQ name(SB) + field offset. Mirror cstage
+		// cg_dotbase_addr's global value-struct arm.
+		emitline("\tLEAQ\t");
+		emitsymname(c, inner.str);
+		emitline("(SB), ");
+		emitline(dstreg);
+		emitline("\n");
+		if (foff != 0) {
+			emitline("\tADDQ\t$");
+			emitint(foff);
+			emitline(", ");
+			emitline(dstreg);
+			emitline("\n");
+		};
 	} else {
 		emitline("\tLEAQ\t");
 		emitoff(innoff + foff);

selfhost/cmd/wwdump/main.combined.ww

@@ -20025,6 +20025,7 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 	// 8 bytes as if it were a pointer-var — wrong shape (cstage
 	// sister fix in cg_dotbase_addr).
 	let lc: *local = localfindnode(c, inner.str);
+	let isglobal: bool = false;
 	if (lc == nil) {
 		let gt: *node = letvartnode(c, base.str);
 		if (gt != nil && gt.kind == nkind.N_TARRAY) {
@@ -20035,7 +20036,12 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 			emitline("\n");
 			return true;
 		};
-		return false;
+		// #249 (sibling of #135): inner is a module-GLOBAL struct value
+		// (let/def), not a local — lc is nil but inner.type_ is a valid
+		// struct. Resolve the field below and emit a global base (LEAQ
+		// name(SB)). A non-struct inner (e.g. an SK_USE module qualifier,
+		// type ty_err) falls through the struct gate to `return false`.
+		isglobal = true;
 	};
 	let bu: *tinfo = inner.type_: *tinfo;
 	for (bu != nil && bu.kind == tykind.TY_NAMED) { bu = bu.under; };
@@ -20072,7 +20078,8 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 	for (ft != nil && ft.kind == tykind.TY_NAMED) { ft = ft.under; };
 	if (ft == nil) { return false; };
 	if (ft.kind != tykind.TY_ARRAY) { return false; };
-	let innoff: i64 = lc.off: i64;
+	let innoff: i64 = 0;
+	if (lc != nil) { innoff = lc.off: i64; };
 	if (viaptr) {
 		emitline("\tMOVQ\t");
 		emitoff(innoff);
@@ -20086,6 +20093,21 @@ fn dotbaseaddr(c: *cgen, base: *node, dstreg: str) bool = {
 			emitline(dstreg);
 			emitline("\n");
 		};
+	} else if (isglobal) {
+		// #249: LEAQ name(SB) + field offset. Mirror cstage
+		// cg_dotbase_addr's global value-struct arm.
+		emitline("\tLEAQ\t");
+		emitsymname(c, inner.str);
+		emitline("(SB), ");
+		emitline(dstreg);
+		emitline("\n");
+		if (foff != 0) {
+			emitline("\tADDQ\t$");
+			emitint(foff);
+			emitline(", ");
+			emitline(dstreg);
+			emitline("\n");
+		};
 	} else {
 		emitline("\tLEAQ\t");
 		emitoff(innoff + foff);

test/wcc/949_structlit_arrfield_run.c

@@ -0,0 +1,185 @@
+/*
+ * 949_structlit_arrfield_run — runtime + byte-id net for #249: struct
+ * array-field init/read silent miscompiles. Two distinct roots, both
+ * gate-blind (cs==ww broken identically pre-fix):
+ *
+ *   BUG B — reading an array field of a module-GLOBAL struct value
+ *   (`G.arr[i]`). The N_INDEX fallback's cg_dotbase_addr / dotbaseaddr
+ *   helper (the #135 sibling) had no module-global-struct base arm:
+ *   cstage emitted `LEAQ (BP)` (read the stack → 0), wwstage fell to
+ *   cgexpr(N_DOT) which loaded the field VALUE as a pointer → SEGFAULT.
+ *   The .data was already correct (emit_struct_lit_bytes #129 A.3); only
+ *   the READ base address was wrong. Fix: a global value-struct base
+ *   emits `LEAQ name(SB) (+ ADDQ field_off)`, mirroring the scalar
+ *   global-field read (cgen.c:7532). const globals are def_isstructdef.
+ *
+ *   BUG A — initializing an array field from a struct literal
+ *   (`e{ arr = [..] }`) as a local. cg_structlit_fill / cgstructlitfill
+ *   had no TY_ARRAY field arm; the array field fell to the generic
+ *   scalar tail (cgexpr the N_ARRLIT → AX, store one sized word) which
+ *   silently DROPPED every element. Fix: a TY_ARRAY field arm element-
+ *   wise stores the N_ARRLIT at base+field_off+i*esz, reusing the proven
+ *   N_LET array-init shape. The GLOBAL literal-init path is unaffected
+ *   (it goes through emit_struct_lit_bytes, already correct).
+ *
+ * cstage `ww build` + run for exit code; w6c vs w6c_ww `.s` cmp for the
+ * rule-10 byte-id gate. u8 element rows only (a `[N]u8` read cast to i32
+ * is byte-id; wider widths hit the i32-return MOVL/MOVSXD cs/ww
+ * divergence — see 949_dotbase_arr_run's deferral note).
+ */
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/stat.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;
+}
+
+struct row { const char *label; const char *src; int want_exit; };
+
+static const struct row rows[] = {
+	/* BUG B: read [4]u8 field of a module-global `let` struct at idx 0
+	 * (field_off 0). Pre-fix cstage→0, wwstage→SEGFAULT. encmap[0]='A'
+	 * (65). */
+	{ "global_let_read",
+	  "package main;\n"
+	  "type e = struct { encmap: [4]u8 };\n"
+	  "let g: e = e { encmap = [65u8, 66u8, 67u8, 68u8] };\n"
+	  "export fn main() i32 = { return g.encmap[0]: i32; };\n", 65 },
+	/* BUG B: read [4]u8 field of a module-global `def` (const) struct
+	 * at idx 2, with a non-zero field offset (a [3]u8 pad ahead of it) —
+	 * exercises def_isstructdef + the ADDQ $field_off arm. The base64
+	 * `const std_encoding` shape. pad ahead, encmap[2]='C' (67). */
+	{ "global_def_read_off",
+	  "package main;\n"
+	  "type e = struct { pad: [3]u8, encmap: [4]u8 };\n"
+	  "def G: e = e { pad = [9u8, 9u8, 9u8],"
+	  " encmap = [65u8, 66u8, 67u8, 68u8] };\n"
+	  "export fn main() i32 = { return G.encmap[2]: i32; };\n", 67 },
+	{ NULL, NULL, 0 }
+};
+
+static int
+slurp_eq(const char *a, const char *b)
+{
+	FILE *fa = fopen(a, "rb");
+	FILE *fb = fopen(b, "rb");
+	if (!fa || !fb) { if (fa) fclose(fa); if (fb) fclose(fb); return -1; }
+	int rc = 0;
+	for (;;) {
+		int ca = fgetc(fa);
+		int cb = fgetc(fb);
+		if (ca != cb) { rc = -1; break; }
+		if (ca == EOF) break;
+	}
+	fclose(fa); fclose(fb);
+	return rc;
+}
+
+int
+main(void)
+{
+	const char *bin = getenv("BIN");
+	if (!bin) bin = "out/bin";
+	char absbin[1024];
+	if (bin[0] != '/') {
+		char cwd[1024];
+		if (getcwd(cwd, sizeof cwd) == NULL) return 1;
+		snprintf(absbin, sizeof absbin, "%s/%s", cwd, bin);
+		bin = absbin;
+	}
+
+	char w6c[1100], w6c_ww[1100];
+	snprintf(w6c, sizeof w6c, "%s/w6c", bin);
+	snprintf(w6c_ww, sizeof w6c_ww, "%s/w6c_ww", bin);
+	if (access(w6c_ww, X_OK) != 0) {
+		fprintf(stderr, "structlit_arrfield: w6c_ww missing — cannot run "
+		    "the cs==ww byte-id gate (the whole point of this test)\n");
+		return 1;
+	}
+
+	int n = 0, fail = 0;
+	for (int i = 0; rows[i].src; i++, n++) {
+		char src[64];
+		snprintf(src, sizeof src, "/tmp/wwsaf_%d_%d.ww", getpid(), i);
+		FILE *f = fopen(src, "wb");
+		if (f == NULL) { fail++; continue; }
+		fputs(rows[i].src, f);
+		fclose(f);
+
+		char tmpdir[64];
+		snprintf(tmpdir, sizeof tmpdir, "/tmp/wwsaf_%d_d_%d",
+		    getpid(), i);
+		mkdir(tmpdir, 0755);
+
+		char cmd[2048];
+		snprintf(cmd, sizeof cmd, "cd %s && %s/ww build %s",
+		    tmpdir, bin, src);
+		if (runwait(cmd) != 0) {
+			fprintf(stderr, "row[%s]: cstage build failed\n",
+			    rows[i].label);
+			fail++;
+			unlink(src); rmdir(tmpdir);
+			continue;
+		}
+
+		char outbin[128];
+		const char *base = strrchr(src, '/');
+		base = base ? base + 1 : src;
+		snprintf(outbin, sizeof outbin, "%s/%s", tmpdir, base);
+		char *dot = strrchr(outbin, '.');
+		if (dot && strcmp(dot, ".ww") == 0) *dot = '\0';
+
+		int got = runwait(outbin);
+		if (got != rows[i].want_exit) {
+			fprintf(stderr, "row[%s]: cstage exit %d, want %d\n",
+			    rows[i].label, got, rows[i].want_exit);
+			fail++;
+		}
+		unlink(outbin); rmdir(tmpdir);
+
+		char cs_s[64], ws_s[64];
+		snprintf(cs_s, sizeof cs_s, "/tmp/wwsaf_%d_%d_cs.s",
+		    getpid(), i);
+		snprintf(ws_s, sizeof ws_s, "/tmp/wwsaf_%d_%d_ww.s",
+		    getpid(), i);
+
+		snprintf(cmd, sizeof cmd, "%s -o %s %s 2>/dev/null",
+		    w6c, cs_s, src);
+		if (runwait(cmd) != 0) {
+			fprintf(stderr, "row[%s]: w6c failed\n", rows[i].label);
+			fail++; unlink(src); continue;
+		}
+		snprintf(cmd, sizeof cmd, "%s -o %s %s 2>/dev/null",
+		    w6c_ww, ws_s, src);
+		if (runwait(cmd) != 0) {
+			fprintf(stderr, "row[%s]: w6c_ww failed\n",
+			    rows[i].label);
+			fail++; unlink(src); unlink(cs_s); continue;
+		}
+		if (slurp_eq(cs_s, ws_s) != 0) {
+			fprintf(stderr,
+			    "row[%s]: cstage/wwstage .s DIFFER (rule-10 "
+			    "byte-id violation)\n", rows[i].label);
+			fail++;
+		}
+		unlink(src); unlink(cs_s); unlink(ws_s);
+	}
+
+	if (fail) {
+		fprintf(stderr, "%d/%d structlit-arrfield tests failed\n",
+		    fail, n);
+		return 1;
+	}
+	printf("structlit_arrfield: %d/%d ok (cstage run + cs==ww byte-id)\n",
+	    n, n);
+	return 0;
+}