Hojun-Cho/ww
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

w6c+wwstage: construct + bind tuple-in-union payload (#242)

Browse Source 
A mixed-scalar tuple WRAPPED IN A TAGGED UNION (the (neg, n) shape Hare's
strconv parseint returns, ((bool,u64)|invalid|overflow)) miscompiled three
ways, all gate-blind (no bootstrap tuple-in-union):

(a) cstage CONSTRUCTION: a tuple variant fell through the N_RETURN scalar
    shuffle, which ZEROED tag + payload — the operands were never packed.
    Route the tuple variant through the scratch-slot widen path; add a
    TY_TUPLE arm to cg_widen_tagged_store that packs each element into the
    union payload at the register-ABI 8B stride + sets the variant tag.

(b) wwstage CHECKER: `let (a,b)=t` over a plain tuple ident (the match-
    bound union payload) left the un-annotated binders UNTYPED, so the bin
    node reading them was untyped -> asserttyped abort. The element-type
    distribution only fired for an N_CALL rhs. Consume the rhs tuple type
    for ANY rhs (mirror cstage check.c:2017).

(c) BOTH stages DESTRUCTURE: the register-cursor receive assumes the rhs
    left every element in AX/DX/CX (a call's tuple-return ABI). For a tuple
    IDENT cgexpr loads only word0->AX, so the 2nd binder read a STALE DX.
    Copy each element from the ident's slot at the 8B stride.

Construction is correct at ANY variant position (the resolved tag, not a
default 0); wwstage resolves it via the typeeq core (flatvariantidxt), not
taggedvariantindext whose str/slice shape-fallback would mask a mismatch.

Two rule-7 loud-stops cover shapes this slotted packing can't yet handle,
on BOTH stages, so neither silently miscompiles:

  - a tuple with a SysV-eightbyte-sharing narrow pair (e.g. (i32,i32,u64)),
    caught by the 8+payload > slot-size guard (the eightbyte tuple
    classification is #243);

  - a tuple built from a BARE LITERAL element (`true`/`false`, suffix-less
    `7`). cstage's cg_tag_for_variant can't type the literal (#241), returns
    -1, and loud-stops. wwstage types `true` as bool and `7` as untyped_int,
    so flatvariantidxt WOULD resolve the variant — a program cstage rejects
    but wwstage accepts is the cs!=ww divergence rule 10 forbids. wwstage
    mirrors cstage's CONDITION (a bare-literal element), not its -1
    mechanism, with an explicit guard that aligns the richer side DOWN. Lift
    BOTH guards together when #241 lands cstage literal typing -> symmetric
    accept.

Test 940_tuple_in_union: 4 K_RUN rows (variant 0, void arm, tuple at
variant 1 two ways) x cstage-run + wwstage-run + cs==ww byte-id, plus 2
K_BUILDERR rows (eightbyte-share, bare-literal) asserting a loud stop with
the #242 diagnostic on BOTH drivers = 16 ok.
This commit is contained in:
Hojun-Cho
2026-06-01 19:56:52 +09:00
parent b79f005489
commit 6fc85f9aaf
8 changed files with 1114 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

78
selfhost/cmd/wcc/cgenstmt.ww
View File

@@ -265,7 +265,13 @@ fn cgreturn(c: *cgen, n: *node) void = {
// INTEGER cursor (ref/qbe/amd64/sysv.c retr L95-108). Both rows
// loud-stop at their cap (rule-7): INTEGER 4, SSE 2. The SAME
// class split drives the receive sites.
if (rhs.kind == nkind.N_TUPLE) {
// #242: a bare tuple return packs into the register cursor; a
// tuple WRAPPED IN A TAGGED UNION must instead pack into the
// union payload (tag + words) — fall through to the tagged path
// below, which routes it via cgwidentaggedstore. Without this
// guard the bare-tuple arm fired first and dropped the tag,
// returning (AX=word0, DX=word1) with no tag word.
if (rhs.kind == nkind.N_TUPLE && !istaggedtype(c, c.fnret)) {
let ssecap: i32 = TUPLE_SSECAP; // X0,X1 per SysV
let gptotal: i32 = 0;
let ssecount: i32 = 0;
@@ -469,6 +475,11 @@ fn cgreturn(c: *cgen, n: *node) void = {
if (rhstaggedident(c, rhs) != nil) {
needswiden = true;
};
// #242: a tuple variant packs into the union
// payload via cgwidentaggedstore's TY_TUPLE arm.
if (rhs.kind == nkind.N_TUPLE) {
needswiden = true;
};
};
};
if (needswiden) {
@@ -1883,6 +1894,71 @@ fn cgmlet(c: *cgen, n: *node) void = {
let sretrecv: i32 = 0;
if (rhs.kind == nkind.N_CALL) { sretrecv = callsretsize(c, rhs); };
// #242: rhs is a tuple already materialised in a local slot (a match-
// bound union payload, `let (a,b)=t`), NOT a register-returning call.
// cgexpr(tuple ident) loads only word0->AX, so the register cursor
// path below reads DX/CX stale. Copy each element from the ident's
// slot at the register-ABI 8B stride (24B for a slice/str header) —
// the SAME layout the tagged construct + match payload-bind write.
// Mirror of cstage cgen.c N_MLET tuple-ident arm. The binding element
// types ride l.lhs (stamped by the checker's stamptuplebinds).
if (rhs.kind == nkind.N_IDENT) {
let rl: *local = localfindnode(c, rhs.str);
if (rl != nil) {
let rti: *tinfo = rl.tnode.type_: *tinfo;
for (rti != nil && rti.kind == tykind.TY_NAMED) { rti = rti.under; };
if (rti != nil) { if (rti.kind == tykind.TY_TUPLE) {
let srcoff: i32 = rl.off;
let foff: i32 = 0;
let lb: *node = n.list;
for (lb != nil) {
let tn: *node = lb.lhs;
let isflt: bool = isfloattype(c, tn);
let wide: bool = isstrtype(c, tn) || isslicetype(c, tn);
let esz: i32 = 8;
let eti: *tinfo = nil;
if (tn != nil) { eti = tn.type_: *tinfo; };
if (eti != nil) { esz = eti.size: i32; };
let bsz: i32 = 8;
if (wide) { bsz = tyslicesize(): i32; };
let off: i32 = localadd(c, lb.str, bsz, tn);
if (isflt) {
let mov: str = "MOVSD";
if (isf32type(c, tn)) { mov = "MOVSS"; };
emitline("\t"); emitline(mov); emitline("\t");
emitoff((srcoff + foff): i64);
emitline("(BP), X0\n");
emitline("\t"); emitline(mov); emitline("\tX0, ");
emitoff(off: i64); emitline("(BP)\n");
} else { if (wide) {
let k: i32 = 0;
for (k < esz) {
emitline("\tMOVQ\t");
emitoff((srcoff + foff + k): i64);
emitline("(BP), AX\n");
emitline("\tMOVQ\tAX, ");
emitoff((off + k): i64);
emitline("(BP)\n");
k += 8;
};
} else {
let lop: str = tnodeloadop(c, tn, esz);
let sop: str = tnodestoreop(c, tn, esz);
emitline("\t"); emitline(lop); emitline("\t");
emitoff((srcoff + foff): i64);
emitline("(BP), AX\n");
emitline("\t"); emitline(sop); emitline("\tAX, ");
emitoff(off: i64); emitline("(BP)\n");
}; };
if (wide) { foff += 24; } else { foff += 8; };
lb = lb.next;
};
c.lastwasreturn = 0;
return;
}; };
};
};
cgexpr(c, rhs);
if (sretrecv > 0) {