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

selfhost: fix several wwstage cgen miscompilations

Browse Source 
Surfaced via examples/lisp, which had to work around the following in
source. Each lowering now matches cstage on the same shape.

- cgassign / cgdot: two-level field through a non-pointer sub-struct.
  `(*L).cur.kind = k` (cur a struct-by-value field of L) silently
  dropped the store; the corresponding read fell into the SB-symbol
  fallback and the linker reported `undefined reference to kind`. The
  two new branches resolve outer-field offset + inner-field offset
  and emit a single direct store/load at the combined slot, both for
  T-by-value and *T-base shapes.

- cgdot: `xs[i].field` chains the trailing field load through the
  N_INDEX result for [N]T / []T / *T element-of-struct-ptr. The
  cgforrange loop variable now carries the elem tnode so the same
  fast path covers `for (let x .. xs) { x.field }`.

- cgindex / cgassign: top-level `[N]T` array and `*T` pointer used
  as an index base. cgindex now emits LEAQ name(SB) (array) or
  MOVQ name(SB) (pointer) with the correct element scaling; without
  this the fallback emitted neither base and walked off the saved
  BP slot. Adds letvartnode() helper, an N_TARRAY branch to
  letemitsize so the array shows up in c.lets, and an N_TARRAY
  initialiser path in emitletdataw that lays the literal bytes into
  DATAW.

- cglet / scanlocals: infer the local's tnode for an unannotated
  `let x = f()` / `let x = f()?`. inferletcalltype() reads the
  callee's declared return; `?` and `!` strip to the success variant
  so a tagged-union let allocates the full 24B slot and the
  struct-field dispatch in cgdot/cgassign sees the right type.
  letslotsize now defers to slotsize on the inferred type.

- slotsize: follow type aliases for tagged-union variants. With
  `type parserr = !str;`, the variant slot was 8B instead of the
  required 16B; the tagged let stomped on the next slot at the
  AX/DX/CX spill.

- cgreturn: tagged-union return forwarding. `return f();` where f
  also returns a tagged union now passes the (tag, payload1,
  payload2) triple through unchanged instead of re-wrapping it.

- cgreturn / cglet / taggedvariantindex: dispatch by variant name
  with module-qualified-vs-bare matching, and recognise N_STRUCTLIT
  as the variant tag for `return eof{};`. cgexpr default emits
  `MOVQ $0, AX` so the surrounding return shuffle isn't left with
  a stale AX.

- isstrtype / nodeisstr: resolve through `!T` aliases. `parserr =
  !str` was not propagating the str-shape to the rhs check and the
  MOVQ BX,CX shuffle was being dropped from str-typed local
  returns.

- exprfloatkind: recognise `p.field` as f64/f32 when the struct
  field is so declared, so `v.fval: i64` lowers to CVTTSD2SI on X0.

- cgassign: str field on a direct struct local writes both halves.
  `L.src = s;` previously dropped s.len.

- cgcall: pop into the int reg window only up to 6 (DI..R9); rest
  stays on the stack and the caller emits ADDQ to clean up.
  cgfnparams accepts >6-arg signatures by registering the overflow
  params at positive BP offsets (16+8*k(BP)), no spill instruction
  emitted.

All 26 harness tests pass; bootstrap reaches a byte-stable fixed
point at ww3 == ww4.
This commit is contained in:
Hojun-Cho
2026-05-13 03:06:46 +09:00
parent ebfd8c3652
commit 7c75dd218a
5 changed files with 854 additions and 108 deletions
Download Patch File Download Diff File Expand all files Collapse all files

110
selfhost/cmd/wcc/cgen.ww
View File

@@ -381,6 +381,20 @@ fn localalloc(c: *cgen, name: str, sz: i32, tnode: *node) i32 = {
return off;
};
// localaddstack — register a param at a positive BP offset. Used for
// args that overflow the 6 SysV int / 8 float reg windows; the caller
// pushes them in reverse, so each spilled arg lives at 16(BP), 24(BP),
// etc. (after the saved RIP+BP). No spill instruction is emitted; the
// slot IS the caller's stack slot.
fn localaddstack(c: *cgen, name: str, tnode: *node, off: i32) void = {
let l: *local = amalloc(c.a, 48u64): *local;
l.name = name;
l.off = off;
l.tnode = tnode;
l.lnext = c.locals;
c.locals = l;
};
fn localadd(c: *cgen, name: str, sz: i32, tnode: *node) i32 = {
// Name-based slot reuse for N_LETs and params: if `name` is
// already declared in this function, return its existing
@@ -634,6 +648,22 @@ fn letemitsize(c: *cgen, d: *node) i32 = {
for (t != nil) {
if (t.kind == nkind.N_TPTR) { return 8; };
if (t.kind == nkind.N_TSLICE) { return 24; };
if (t.kind == nkind.N_TARRAY) {
let lenn: *node = t.rhs;
let elemn: *node = t.lhs;
let alen: i32 = 1;
if (lenn != nil) {
if (lenn.kind == nkind.N_INTLIT) { alen = lenn.uval: i32; };
};
let esz: i32 = 8;
if (elemn != nil) {
if (elemn.kind == nkind.N_TNAME) {
let ps: i32 = primsize(elemn.str);
if (ps > 0) { esz = ps; };
};
};
return alen * esz;
};
if (t.kind != nkind.N_TNAME) { return 0; };
let nm: str = t.str;
if (letscalarprim(nm)) { return 8; };
@@ -683,6 +713,19 @@ fn isletvar(c: *cgen, name: str) bool = {
// aliases to mirror C cgen's `let_isstr`. Used by cgident/cgdot/
// cgassign to pick the (LEAQ, MOVQ, MOVQ) sequence over the bare
// MOVQ scalar load.
// letvartnode — direct lookup of a top-level let's tnode. Used by
// cgindex / cgassign to detect global `[N]T` arrays and `*T`
// pointers, where the addressing path needs LEAQ name(SB) (array)
// or MOVQ name(SB) (pointer) and the element size from T.
fn letvartnode(c: *cgen, name: str) *node = {
let lv: *letvar = c.lets;
for (lv != nil) {
if (streq(lv.name, name)) { return lv.tnode; };
lv = lv.lvnext;
};
return nil;
};
fn letvarisstr(c: *cgen, name: str) bool = {
let lv: *letvar = c.lets;
for (lv != nil) {
@@ -1055,6 +1098,73 @@ fn emitletdataw(c: *cgen, file: *node) void = {
emitline("\"\n");
};
};
// Top-level `[N]T = [a, b, ...]` array global.
// Emits N*esz bytes with each element's bytes
// little-endian for the declared primitive width.
// Without this, `let arr: [N]T = ...` references
// from function bodies link-fail with `undefined
// reference to arr`, and bare-name addressing
// (LEAQ arr(SB)) inside cgindex / cgassign has no
// symbol to bind to.
if (d.lhs != nil) {
if (d.lhs.kind == nkind.N_TARRAY) {
let elemn: *node = d.lhs.lhs;
let esz: i32 = 8;
if (elemn != nil) {
if (elemn.kind == nkind.N_TNAME) {
let ps: i32 = primsize(elemn.str);
if (ps > 0) { esz = ps; };
};
};
let total: i32 = sz;
let alen: i32 = total / esz;
let elems: *node = nil;
if (d.rhs != nil) {
if (d.rhs.kind == nkind.N_ARRLIT) {
elems = d.rhs.list;
};
};
emitline("DATAW ");
emitsymname(c, nm);
emitline("(SB),\"");
let i: i32 = 0;
let e: *node = elems;
let fillv: u64 = 0u64;
let inrepeat: bool = false;
for (i < alen) {
let v: u64 = fillv;
if (!inrepeat && e != nil) {
if (e.kind == nkind.N_FIELD) {
if (streq(e.str, "...")) {
// `..., ...` repeat marker: prior v stays.
inrepeat = true;
} else {
if (e.lhs != nil) {
if (e.lhs.kind == nkind.N_INTLIT) { v = e.lhs.uval; };
if (e.lhs.kind == nkind.N_RUNELIT) { v = e.lhs.uval; };
};
fillv = v;
e = e.next;
};
} else {
if (e.kind == nkind.N_INTLIT) { v = e.uval; };
if (e.kind == nkind.N_RUNELIT) { v = e.uval; };
fillv = v;
e = e.next;
};
};
let nb: u64 = v;
let b: i32 = 0;
for (b < esz) {
emitdatawbyte((nb & 255u64): u8);
nb = nb >> 8u64;
b += 1;
};
i += 1;
};
emitline("\"\n");
};
};
};
};
d = d.next;

37
selfhost/cmd/wcc/cgendecl.ww
View File

@@ -167,6 +167,11 @@ fn cgfnparams(c: *cgen, params: *node) void = {
let p: *node = params;
let idx: i32 = 0;
let fidx: i32 = 0;
// Cursor for args that overflow the SysV reg windows. Each
// stack-passed arg lives at 16+8*k(BP) — no spill, the local
// is registered with a *positive* offset pointing into the
// caller's frame. Mirrors C cgen's cg_stack_arg_cursor.
let stkcursor: i32 = 0;
for (p != nil) {
if (p.kind == nkind.N_PARAM) {
let nm: str = p.str;
@@ -175,6 +180,7 @@ fn cgfnparams(c: *cgen, params: *node) void = {
// (X0..X7). 8B (f64) or 4B (f32) slot.
let fsz: i32 = 8;
if (isf32type(c, p.lhs)) { fsz = 4; };
if (fidx < 8) {
let off: i32 = localadd(c, nm, fsz, p.lhs);
let mov: str = "MOVSD";
if (fsz == 4) { mov = "MOVSS"; };
@@ -186,15 +192,18 @@ fn cgfnparams(c: *cgen, params: *node) void = {
emitoff(off: i64);
emitline("(BP)\n");
fidx += 1;
} else {
localaddstack(c, nm, p.lhs, 16 + stkcursor*8);
stkcursor += 1;
};
p = p.next;
continue;
};
if (istaggedtype(p.lhs)) {
// tagged-union param: spill size/8 registers
// (tag + value words). Slot sized to match.
let slot: i32 = slotsize(c, p.lhs);
let off: i32 = localadd(c, nm, slot, p.lhs);
let nw: i32 = slot / 8;
if (idx + nw <= 6) {
let off: i32 = localadd(c, nm, slot, p.lhs);
let w: i32 = 0;
for (w < nw) {
emitline("\tMOVQ\t");
@@ -205,8 +214,12 @@ fn cgfnparams(c: *cgen, params: *node) void = {
idx += 1;
w += 1;
};
} else {
localaddstack(c, nm, p.lhs, 16 + stkcursor*8);
stkcursor += nw;
};
} else { if (isslicetype(c, p.lhs)) {
// slice param: 3 regs (ptr, len, cap), 24-byte slot.
if (idx + 3 <= 6) {
let off: i32 = localadd(c, nm, 24, p.lhs);
emitline("\tMOVQ\t");
emitline(argregname(idx));
@@ -226,9 +239,12 @@ fn cgfnparams(c: *cgen, params: *node) void = {
emitoff((off + 16): i64);
emitline("(BP)\n");
idx += 1;
} else {
localaddstack(c, nm, p.lhs, 16 + stkcursor*8);
stkcursor += 3;
};
} else { if (isstrtype(c, p.lhs)) {
// str param: passed in two regs (ptr, len).
// Slot is 16 bytes; ptr at off+0, len at off+8.
if (idx + 2 <= 6) {
let off: i32 = localadd(c, nm, 16, p.lhs);
emitline("\tMOVQ\t");
emitline(argregname(idx));
@@ -243,6 +259,11 @@ fn cgfnparams(c: *cgen, params: *node) void = {
emitline("(BP)\n");
idx += 1;
} else {
localaddstack(c, nm, p.lhs, 16 + stkcursor*8);
stkcursor += 2;
};
} else {
if (idx < 6) {
let off: i32 = localadd(c, nm, 8, p.lhs);
emitline("\tMOVQ\t");
emitline(argregname(idx));
@@ -250,6 +271,10 @@ fn cgfnparams(c: *cgen, params: *node) void = {
emitoff(off: i64);
emitline("(BP)\n");
idx += 1;
} else {
localaddstack(c, nm, p.lhs, 16 + stkcursor*8);
stkcursor += 1;
};
};};};
};
p = p.next;

443
selfhost/cmd/wcc/cgenexpr.ww
View File

@@ -100,6 +100,12 @@ fn cgexpr(c: *cgen, n: *node) void = {
if (k == nkind.N_TRYUNW) { cgtryunw(c, n); return; };
if (k == nkind.N_TYPETEST) { cgtypetest(c, n); return; };
if (k == nkind.N_TYPEASSERT) { cgtypeassert(c, n); return; };
// Default fallback: produce a deterministic AX = 0. Mirrors
// the C cgen's `default: cgexpr_int(c, 0)` branch, which is
// what `return eof{};` (N_STRUCTLIT with an empty !void
// variant) silently relies on — without this AX carries a
// stale value into the tagged-union return shuffle.
emitline("\tMOVQ\t$0, AX\n");
};
// cgtagvariantidx — find the 0-based variant index of `vt` inside the
@@ -524,6 +530,14 @@ fn cgindex(c: *cgen, n: *node) void = {
let esz: i32 = 8;
let signed_elem: bool = false;
let baselocal: *local = nil;
// Global `[N]T` array or `*T` pointer used as an index base.
// The local-ident lookup above misses it; we need LEAQ name(SB)
// (array, the symbol IS the storage) or MOVQ name(SB) (pointer,
// the symbol holds the address) to feed the addend.
let isglobalarr: bool = false;
let isglobalptr: bool = false;
let globalname: str;
globalname.ptr = nil; globalname.len = 0;
if (base != nil) {
if (base.kind == nkind.N_IDENT) {
let bn: str = base.str;
@@ -531,6 +545,22 @@ fn cgindex(c: *cgen, n: *node) void = {
if (baselocal != nil) {
esz = elemsizeof(baselocal.tnode);
signed_elem = elemissigned(baselocal.tnode);
} else {
let tn: *node = letvartnode(c, bn);
if (tn != nil) {
if (tn.kind == nkind.N_TARRAY) {
isglobalarr = true;
globalname = bn;
esz = elemsizeof(tn);
signed_elem = elemissigned(tn);
};
if (tn.kind == nkind.N_TPTR) {
isglobalptr = true;
globalname = bn;
esz = elemsizeof(tn);
signed_elem = elemissigned(tn);
};
};
};
} else { if (base.kind == nkind.N_DOT) {
esz = indexbaseesz(c, base);
@@ -543,6 +573,31 @@ fn cgindex(c: *cgen, n: *node) void = {
emitline(", CX\n");
emitline("\tIMULQ\tCX, AX\n");
};
if (isglobalarr || isglobalptr) {
if (isglobalarr) {
emitline("\tLEAQ\t");
emitsymname(c, globalname);
emitline("(SB), BX\n");
} else {
emitline("\tMOVQ\t");
emitsymname(c, globalname);
emitline("(SB), BX\n");
};
emitline("\tADDQ\tAX, BX\n");
if (esz == 16) {
emitline("\tMOVQ\t8(BX), CX\n");
emitline("\tMOVQ\t(BX), AX\n");
emitline("\tMOVQ\tCX, BX\n");
return;
};
if (esz == 1) { emitline("\tMOVZBQ\t(BX), AX\n"); }
else { if (esz == 4) {
if (signed_elem) { emitline("\tMOVSXD\t(BX), AX\n"); }
else { emitline("\tMOVL\t(BX), AX\n"); };
}
else { emitline("\tMOVQ\t(BX), AX\n"); };};
return;
};
if (baselocal != nil) {
let tn: *node = baselocal.tnode;
let isarray: bool = false;
@@ -782,7 +837,7 @@ fn cgmatch(c: *cgen, n: *node) void = {
let found: bool = false;
for (v != nil) {
if (v.kind == nkind.N_TNAME) {
if (streq(v.str, patname)) {
if (variantnamematch(v.str, patname)) {
want = idx;
found = true;
v = nil;
@@ -1210,6 +1265,71 @@ fn cgdot(c: *cgen, n: *node) void = {
};
};
};
// `xs[i].field` — slice/array/ptr-of-struct element field access.
// Without this the cgen falls through to the module-qualified
// SB fallback below and emits `MOVQ <fld>(SB), AX` (linker
// reports `undefined reference to <fld>`). cgexpr(c, lhs)
// dispatches to cgindex which leaves the element value in AX
// — for a []*T element that's the *T pointer, so we just chain
// the field load through (AX).
if (lhs != nil) {
if (lhs.kind == nkind.N_INDEX) {
let idxbase: *node = lhs.lhs;
if (idxbase != nil) { if (idxbase.kind == nkind.N_IDENT) {
let lc: *local = localfindnode(c, idxbase.str);
if (lc != nil) { if (lc.tnode != nil) {
let tn: *node = lc.tnode;
let elemt: *node = nil;
let tk: nkind = tn.kind;
if (tk == nkind.N_TSLICE) { elemt = tn.lhs; };
if (tk == nkind.N_TARRAY) { elemt = tn.lhs; };
if (tk == nkind.N_TPTR) { elemt = tn.lhs; };
if (elemt != nil) { if (elemt.kind == nkind.N_TPTR) {
let inner: *node = elemt.lhs;
if (inner != nil) { if (inner.kind == nkind.N_TNAME) {
let sname: str = inner.str;
let si: *structinfo = structlookup(c, sname);
if (si != nil) {
let fi: *fieldinfo = si.fields;
for (fi != nil) {
if (streq(fi.fname, fld)) {
cgexpr(c, lhs); // AX = *Struct
if (isstrtype(c, fi.tnode)) {
emitline("\tMOVQ\t");
emitdispreg((fi.foff + 8): i64, "AX");
emitline(", BX\n");
emitline("\tMOVQ\t");
emitdispreg(fi.foff: i64, "AX");
emitline(", AX\n");
return;
};
if (isfloattype(c, fi.tnode)) {
let mov: str = "MOVSD";
if (isf32type(c, fi.tnode)) { mov = "MOVSS"; };
emitline("\t");
emitline(mov);
emitline("\t");
emitdispreg(fi.foff: i64, "AX");
emitline(", X0\n");
return;
};
let lop: str = fieldloadop(fi);
emitline("\t");
emitline(lop);
emitline("\t");
emitdispreg(fi.foff: i64, "AX");
emitline(", AX\n");
return;
};
fi = fi.finext;
};
};
};};
};};
};};
};};
};
};
// Module-qualified value reference: `mod.name` where `mod`
// is nkind.N_IDENT bound as skind.SK_USE and the leaf isn't a local.
// Treat as a SB symbol — `MOVQ leaf(SB), AX`. Same fallback
@@ -1287,6 +1407,120 @@ fn cgdot(c: *cgen, n: *node) void = {
};
};
};
// Chained `(ident).f1.f2` read where f1 is a struct-by-value
// field. Mirror of the cgassign branch added for the same shape.
// Without this, `L.cur.kind` (cur a by-value struct of *L)
// falls into the SB-fallback and emits `MOVQ kind(SB), AX`.
if (lhs != nil) {
if (lhs.kind == nkind.N_DOT) {
let inner: *node = lhs.lhs;
let innerfld: str = lhs.str;
if (inner != nil) { if (inner.kind == nkind.N_IDENT) {
let lc: *local = localfindnode(c, inner.str);
if (lc != nil) { if (lc.tnode != nil) {
let tn: *node = lc.tnode;
let lkind: nkind = tn.kind;
let outname: str;
outname.ptr = nil; outname.len = 0;
let isptr: bool = false;
if (lkind == nkind.N_TNAME) { outname = tn.str; };
if (lkind == nkind.N_TPTR) {
let pe: *node = tn.lhs;
if (pe != nil) { if (pe.kind == nkind.N_TNAME) {
outname = pe.str;
isptr = true;
};};
};
if (outname.len > 0) {
let osi: *structinfo = structlookup(c, outname);
if (osi != nil) {
let ofi: *fieldinfo = osi.fields;
for (ofi != nil) {
if (streq(ofi.fname, innerfld)) {
let oft: *node = ofi.tnode;
if (oft != nil) { if (oft.kind == nkind.N_TNAME) {
if (primsize(oft.str) == 0) {
let isi: *structinfo = structlookup(c, oft.str);
if (isi != nil) {
let ffi: *fieldinfo = isi.fields;
for (ffi != nil) {
if (streq(ffi.fname, fld)) {
let totoff: i32 = ofi.foff + ffi.foff;
if (isstrtype(c, ffi.tnode)) {
if (isptr) {
emitline("\tMOVQ\t");
emitoff(lc.off: i64);
emitline("(BP), CX\n");
emitline("\tMOVQ\t");
emitdispreg((totoff + 8): i64, "CX");
emitline(", BX\n");
emitline("\tMOVQ\t");
emitdispreg(totoff: i64, "CX");
emitline(", AX\n");
} else {
emitline("\tMOVQ\t");
emitoff((lc.off + totoff): i64);
emitline("(BP), AX\n");
emitline("\tMOVQ\t");
emitoff((lc.off + totoff + 8): i64);
emitline("(BP), BX\n");
};
return;
};
if (isfloattype(c, ffi.tnode)) {
let mov: str = "MOVSD";
if (isf32type(c, ffi.tnode)) { mov = "MOVSS"; };
if (isptr) {
emitline("\tMOVQ\t");
emitoff(lc.off: i64);
emitline("(BP), BX\n");
emitline("\t");
emitline(mov);
emitline("\t");
emitdispreg(totoff: i64, "BX");
emitline(", X0\n");
} else {
emitline("\t");
emitline(mov);
emitline("\t");
emitoff((lc.off + totoff): i64);
emitline("(BP), X0\n");
};
return;
};
let lop: str = fieldloadop(ffi);
if (isptr) {
emitline("\tMOVQ\t");
emitoff(lc.off: i64);
emitline("(BP), BX\n");
emitline("\t");
emitline(lop);
emitline("\t");
emitdispreg(totoff: i64, "BX");
emitline(", AX\n");
} else {
emitline("\t");
emitline(lop);
emitline("\t");
emitoff((lc.off + totoff): i64);
emitline("(BP), AX\n");
};
return;
};
ffi = ffi.finext;
};
};
};
};};
};
ofi = ofi.finext;
};
};
};
};};
};};
};
};
return;
};
@@ -1732,16 +1966,20 @@ fn cgcall(c: *cgen, n: *node) void = {
// SUBQ+MOVSD; pop into the XMM stream (X0..X7). Everything else
// pops into the int stream (DI..R9) per the SysV ABI. Walk the
// args list alongside the pop counter so we know each arg's
// register class.
// register class. SysV has only 6 int arg regs (DI/SI/DX/CX/R8/R9);
// the remaining slots stay on the stack and the callee reads them
// via 16+8*k(BP). Caller-cleanup is emitted after the CALL.
let intidx: i32 = 0;
let fpidx: i32 = 0;
let a: *node = n.list;
let popped: i32 = 0;
let stackslots: i32 = 0;
for (a != nil) {
let fk: i32 = exprfloatkind(c, a);
if (fk != 0) {
let mov: str = "MOVSD";
if (fk == 1) { mov = "MOVSS"; };
if (fpidx < 8) {
emitline("\t");
emitline(mov);
emitline("\t(SP), ");
@@ -1749,26 +1987,27 @@ fn cgcall(c: *cgen, n: *node) void = {
emitline("\n");
emitline("\tADDQ\t$8, SP\n");
fpidx += 1;
} else {
stackslots += 1;
};
popped += 1;
} else {
emitline("\tPOPQ\t");
emitline(argregname(intidx));
emitline("\n");
intidx += 1;
popped += 1;
// Multi-word args (str=2, slice/tagged=3): drain
// the remaining words into successive int regs.
let extra: i32 = 0;
if (nodeisstr(c, a)) { extra = 1; };
if (nodeisslice(c, a)) { extra = 2; };
let e: i32 = 0;
for (e < extra) {
let words: i32 = 1 + extra;
let w: i32 = 0;
for (w < words) {
if (intidx < 6) {
emitline("\tPOPQ\t");
emitline(argregname(intidx));
emitline("\n");
intidx += 1;
} else {
stackslots += 1;
};
popped += 1;
e += 1;
w += 1;
};
};
a = a.next;
@@ -1779,10 +2018,14 @@ fn cgcall(c: *cgen, n: *node) void = {
// case here is identical pre-port behaviour.
let i: i32 = popped;
for (i < nargs) {
if (intidx < 6) {
emitline("\tPOPQ\t");
emitline(argregname(intidx));
emitline("\n");
intidx += 1;
} else {
stackslots += 1;
};
i += 1;
};
let callee: *node = n.lhs;
@@ -1869,6 +2112,14 @@ fn cgcall(c: *cgen, n: *node) void = {
};
emitline("(SB)\n");
};
// Caller cleanup for stack-passed args (args 7+, or any
// overflow past the int/float reg windows). Mirrors C cgen:
// pushed 8 bytes each, ADDQ them off after the CALL.
if (stackslots > 0) {
emitline("\tADDQ\t$");
emitint((stackslots * 8): i64);
emitline(", SP\n");
};
// SysV returns 16-byte aggregates in (AX, DX). Our str
// convention is (AX, BX), so shuffle for str-returning calls.
if (calleename.len > 0) {
@@ -1991,12 +2242,30 @@ fn cgassign(c: *cgen, n: *node) void = {
let idx: *node = lhs.rhs;
let esz: i32 = 8;
let baselocal: *local = nil;
let isglobalarr: bool = false;
let isglobalptr: bool = false;
let globalname: str;
globalname.ptr = nil; globalname.len = 0;
if (base != nil) {
if (base.kind == nkind.N_IDENT) {
let bn: str = base.str;
baselocal = localfindnode(c, bn);
if (baselocal != nil) {
esz = elemsizeof(baselocal.tnode);
} else {
let tn: *node = letvartnode(c, bn);
if (tn != nil) {
if (tn.kind == nkind.N_TARRAY) {
isglobalarr = true;
globalname = bn;
esz = elemsizeof(tn);
};
if (tn.kind == nkind.N_TPTR) {
isglobalptr = true;
globalname = bn;
esz = elemsizeof(tn);
};
};
};
} else { if (base.kind == nkind.N_DOT) {
esz = indexbaseesz(c, base);
@@ -2013,7 +2282,15 @@ fn cgassign(c: *cgen, n: *node) void = {
emitline("\tIMULQ\tCX, AX\n");
};
emitline("\tPUSHQ\tAX\n"); // scaled idx
if (baselocal != nil) {
if (isglobalarr) {
emitline("\tLEAQ\t");
emitsymname(c, globalname);
emitline("(SB), BX\n");
} else { if (isglobalptr) {
emitline("\tMOVQ\t");
emitsymname(c, globalname);
emitline("(SB), BX\n");
} else { if (baselocal != nil) {
let tn: *node = baselocal.tnode;
let isarray: bool = false;
if (tn != nil) { if (tn.kind == nkind.N_TARRAY) { isarray = true; }; };
@@ -2029,7 +2306,7 @@ fn cgassign(c: *cgen, n: *node) void = {
} else {
cgexpr(c, base);
emitline("\tMOVQ\tAX, BX\n");
};
};};};
emitline("\tPOPQ\tAX\n"); // scaled idx
emitline("\tADDQ\tAX, BX\n");
emitline("\tPOPQ\tAX\n"); // value
@@ -2160,6 +2437,20 @@ fn cgassign(c: *cgen, n: *node) void = {
let fn_: str = fi.fname;
if (streq(fn_, fld)) {
cgexpr(c, n.rhs);
// str field: cgexpr left (AX=ptr, BX=len);
// store both halves at +0/+8. Without this,
// `L.src = s` would only write the ptr and
// `L.src.len` would carry whatever was on the
// stack.
if (isstrtype(c, fi.tnode)) {
emitline("\tMOVQ\tAX, ");
emitoff((lc.off + fi.foff): i64);
emitline("(BP)\n");
emitline("\tMOVQ\tBX, ");
emitoff((lc.off + fi.foff + 8): i64);
emitline("(BP)\n");
return;
};
// f64/f32 direct struct local store: route via X0.
if (isfloattype(c, fi.tnode)) {
let mov: str = "MOVSD";
@@ -2436,6 +2727,130 @@ fn cgassign(c: *cgen, n: *node) void = {
};
};
};
// Chained `(ident).f1.f2 = v` where f1 is a struct-by-value
// field. The earlier chained-DOT branch handles f1: *T (deref
// then store). This handles f1: T (in-place sub-struct), which
// would otherwise silently emit no store — lispcore's lexer had
// to flatten `cur.kind`/`cur.ival`/... into top-level fields to
// work around it. Only plain `=` is wired; compound on a by-
// value sub-field hasn't surfaced.
if (lhs != nil) {
if (lhs.kind == nkind.N_DOT) {
let base: *node = lhs.lhs;
let fld: str = lhs.str;
if (base != nil) { if (base.kind == nkind.N_DOT) {
let inner: *node = base.lhs;
let innerfld: str = base.str;
if (inner != nil) { if (inner.kind == nkind.N_IDENT) {
let lc: *local = localfindnode(c, inner.str);
if (lc != nil) { if (lc.tnode != nil) {
let tn: *node = lc.tnode;
let lkind: nkind = tn.kind;
let outname: str;
outname.ptr = nil; outname.len = 0;
let isptr: bool = false;
if (lkind == nkind.N_TNAME) { outname = tn.str; };
if (lkind == nkind.N_TPTR) {
let pe: *node = tn.lhs;
if (pe != nil) { if (pe.kind == nkind.N_TNAME) {
outname = pe.str;
isptr = true;
};};
};
if (outname.len > 0) {
let osi: *structinfo = structlookup(c, outname);
if (osi != nil) {
let ofi: *fieldinfo = osi.fields;
for (ofi != nil) {
if (streq(ofi.fname, innerfld)) {
let oft: *node = ofi.tnode;
if (oft != nil) { if (oft.kind == nkind.N_TNAME) {
if (primsize(oft.str) == 0) {
let isi: *structinfo = structlookup(c, oft.str);
if (isi != nil) {
let ffi: *fieldinfo = isi.fields;
for (ffi != nil) {
if (streq(ffi.fname, fld)) {
if (n.op == tkind.TK_ASSIGN) {
let totoff: i32 = ofi.foff + ffi.foff;
cgexpr(c, n.rhs);
if (isstrtype(c, ffi.tnode)) {
if (isptr) {
emitline("\tMOVQ\t");
emitoff(lc.off: i64);
emitline("(BP), CX\n");
emitline("\tMOVQ\tAX, ");
emitdispreg(totoff: i64, "CX");
emitline("\n");
emitline("\tMOVQ\tBX, ");
emitdispreg((totoff + 8): i64, "CX");
emitline("\n");
} else {
emitline("\tMOVQ\tAX, ");
emitoff((lc.off + totoff): i64);
emitline("(BP)\n");
emitline("\tMOVQ\tBX, ");
emitoff((lc.off + totoff + 8): i64);
emitline("(BP)\n");
};
return;
};
if (isfloattype(c, ffi.tnode)) {
let mov: str = "MOVSD";
if (isf32type(c, ffi.tnode)) { mov = "MOVSS"; };
if (isptr) {
emitline("\tMOVQ\t");
emitoff(lc.off: i64);
emitline("(BP), BX\n");
emitline("\t");
emitline(mov);
emitline("\tX0, ");
emitdispreg(totoff: i64, "BX");
emitline("\n");
} else {
emitline("\t");
emitline(mov);
emitline("\tX0, ");
emitoff((lc.off + totoff): i64);
emitline("(BP)\n");
};
return;
};
let sop: str = fieldstoreop(ffi);
if (isptr) {
emitline("\tMOVQ\t");
emitoff(lc.off: i64);
emitline("(BP), BX\n");
emitline("\t");
emitline(sop);
emitline("\tAX, ");
emitdispreg(totoff: i64, "BX");
emitline("\n");
} else {
emitline("\t");
emitline(sop);
emitline("\tAX, ");
emitoff((lc.off + totoff): i64);
emitline("(BP)\n");
};
return;
};
};
ffi = ffi.finext;
};
};
};
};};
};
ofi = ofi.finext;
};
};
};
};};
};};
};};
};
};
// Local-ident target — plain `=` and the simple compound
// forms (+= -= *= /=); other compounds fall back to
// "evaluate rhs, replace". Mirrors C cgen's IDENT-assign path.

50
selfhost/cmd/wcc/cgenstmt.ww
View File

@@ -143,6 +143,25 @@ fn cgreturn(c: *cgen, n: *node) void = {
// Nullable folded `(*T | void)`: just one word; AX is
// already the pointer (or 0). No shuffle, no tag.
if (istaggedtype(c.fnret)) {
// Forwarding a fallible call: `return f();` where f
// also returns a tagged union. The result is already
// in (AX=tag, DX=v0, CX=v1) — no shuffle, no tag.
// Mirrors the rhsreturnstagged path in cglet and the
// !type_istagged guard in C cgen's N_RETURN.
let forwardtagged: bool = false;
if (rhs.kind == nkind.N_CALL) {
let callee: *node = rhs.lhs;
if (callee != nil) {
let calleename: str;
calleename.ptr = nil; calleename.len = 0;
if (callee.kind == nkind.N_IDENT) { calleename = callee.str; };
if (callee.kind == nkind.N_DOT) { calleename = callee.str; };
if (calleename.len > 0) {
let rt: *node = fnretlookup(c, calleename);
if (istaggedtype(rt)) { forwardtagged = true; };
};
};
};
cgexpr(c, rhs);
if (isnullabletype(c.fnret)) {
emitline("\tMOVQ\tBP, SP\n");
@@ -151,6 +170,13 @@ fn cgreturn(c: *cgen, n: *node) void = {
c.lastwasreturn = 1;
return;
};
if (forwardtagged) {
emitline("\tMOVQ\tBP, SP\n");
emitline("\tPOPQ\tBP\n");
emitline("\tRET\n");
c.lastwasreturn = 1;
return;
};
let idx: i32 = taggedvariantindex(c, c.fnret, rhs);
if (nodeisstr(c, rhs)) {
emitline("\tMOVQ\tBX, CX\n");
@@ -213,7 +239,12 @@ fn cgexprstmt(c: *cgen, n: *node) void = {
fn cglet(c: *cgen, n: *node) void = {
let nm: str = n.str;
let sz: i32 = letslotsize(c, n);
let off: i32 = localadd(c, nm, sz, n.lhs);
// `let x = f()?` has no annotation but the cgen's struct-field
// paths need a tnode to dispatch off. Infer from f's tagged
// success variant — see inferletcalltype.
let tn: *node = n.lhs;
if (tn == nil) { tn = inferletcalltype(c, n.rhs); };
let off: i32 = localadd(c, nm, sz, tn);
if (n.rhs != nil) {
let rhs: *node = n.rhs;
// Tagged-union init: `let r: (T | E) = expr;`.
@@ -222,8 +253,8 @@ fn cglet(c: *cgen, n: *node) void = {
// just spill all three.
// - Otherwise rhs is a bare variant value: pack tag +
// value(s).
if (istaggedtype(n.lhs)) {
let nullable: bool = isnullabletype(n.lhs);
if (istaggedtype(tn)) {
let nullable: bool = isnullabletype(tn);
let rhsreturnstagged: bool = false;
if (rhs.kind == nkind.N_CALL) {
let callee: *node = rhs.lhs;
@@ -266,7 +297,7 @@ fn cglet(c: *cgen, n: *node) void = {
c.lastwasreturn = 0;
return;
};
let tagidx: i32 = taggedvariantindex(c, n.lhs, rhs);
let tagidx: i32 = taggedvariantindex(c, tn, rhs);
if (tagidx < 0) { tagidx = 0; };
if (nodeisstr(c, rhs)) {
emitline("\tMOVQ\tAX, ");
@@ -841,9 +872,9 @@ fn cgforrange(c: *cgen, n: *node) void = {
bind_signed[nbinds] = signf;
let bnm: str = m.str;
if (bnm.len > 0) {
bind_off[nbinds] = localadd(c, bnm, slot_sz, nil);
bind_off[nbinds] = localadd(c, bnm, slot_sz, tp);
} else {
bind_off[nbinds] = localalloc(c, mkscratchname(c, "fr"), slot_sz, nil);
bind_off[nbinds] = localalloc(c, mkscratchname(c, "fr"), slot_sz, tp);
};
field_off += fsz;
nbinds += 1;
@@ -863,9 +894,12 @@ fn cgforrange(c: *cgen, n: *node) void = {
bind_signed[0] = paramissigned(elemt);
};
if (n.str.len > 0) {
bind_off[0] = localadd(c, n.str, slot_sz, nil);
// Register with elem tnode so x.field on a loop
// var resolves through the standard local-typed
// path instead of falling into the SB fallback.
bind_off[0] = localadd(c, n.str, slot_sz, elemt);
} else {
bind_off[0] = localalloc(c, mkscratchname(c, "fr"), slot_sz, nil);
bind_off[0] = localalloc(c, mkscratchname(c, "fr"), slot_sz, elemt);
};
nbinds = 1;
};

182
selfhost/cmd/wcc/cgenutil.ww
View File

@@ -190,13 +190,11 @@ fn nodeisstr(c: *cgen, n: *node) bool = {
let nm: str = n.str;
let lc: *local = localfindnode(c, nm);
if (lc != nil) {
let tn: *node = lc.tnode;
if (tn != nil) {
if (tn.kind == nkind.N_TNAME) {
let tnm: str = tn.str;
if (streq(tnm, "str")) { return true; };
};
};
// Use isstrtype so `!str` aliases (parserr = !str) and
// `type foo = str;` chains resolve through. The bare
// `streq("str", ...)` test missed them and dropped the
// MOVQ BX,CX shuffle on returns of str-aliased locals.
if (isstrtype(c, lc.tnode)) { return true; };
};
return false;
};
@@ -684,11 +682,88 @@ fn primsize(name: str) i32 = {
return 0;
};
// variantnamematch — tagged-union variant names are compared as if
// they'd been alias-resolved. Pattern names can be module-qualified
// (`strconv.invalid` from a `case let e: strconv.invalid =>`),
// while the variant's declared name inside its own module is bare
// (`invalid`). With no checker the cgen can't follow imports, so we
// accept exact match plus suffix-after-`.` on either side. Mirrors
// the C cgen's type_eq, which goes through resolved Type pointers.
fn variantnamematch(vname: str, pname: str) bool = {
if (streq(vname, pname)) { return true; };
// `pname` is qualified, `vname` is bare: drop module prefix.
let i: i32 = 0;
for (i < pname.len) {
if (pname[i] == '.': u8) {
let tail: str;
tail.ptr = pname.ptr + i + 1;
tail.len = pname.len - i - 1;
if (streq(tail, vname)) { return true; };
};
i += 1;
};
// `vname` is qualified, `pname` is bare: same trick in reverse.
let j: i32 = 0;
for (j < vname.len) {
if (vname[j] == '.': u8) {
let tail: str;
tail.ptr = vname.ptr + j + 1;
tail.len = vname.len - j - 1;
if (streq(tail, pname)) { return true; };
};
j += 1;
};
return false;
};
// inferletcalltype — for an annotation-less `let x = expr;`, return
// a usable tnode for cgen's struct-aware paths. Today: `let x =
// f()?` infers x's type from the success variant of f's tagged
// return; without this, x has tnode = nil and `x.field` falls into
// the SB-symbol fallback (linker reports `undefined reference to
// <fieldname>`). We don't infer for plain `let x = f()` yet —
// non-tagged returns don't carry their type back the same way.
fn inferletcalltype(c: *cgen, rhs: *node) *node = {
if (rhs == nil) { return nil; };
// `?` (N_TRYPROP) and `!` (N_TRYUNW) both unwrap a tagged
// return to its success variant; the rhs we want the type of
// is the inner call expression.
let unwrap: bool = false;
let call: *node = rhs;
if (rhs.kind == nkind.N_TRYPROP) { call = rhs.lhs; unwrap = true; };
if (rhs.kind == nkind.N_TRYUNW) { call = rhs.lhs; unwrap = true; };
if (call == nil) { return nil; };
if (call.kind != nkind.N_CALL) { return nil; };
let callee: *node = call.lhs;
if (callee == nil) { return nil; };
let cname: str;
cname.ptr = nil; cname.len = 0;
if (callee.kind == nkind.N_IDENT) { cname = callee.str; };
if (callee.kind == nkind.N_DOT) { cname = callee.str; };
if (cname.len == 0) { return nil; };
let rt: *node = fnretlookup(c, cname);
if (rt == nil) { return nil; };
if (unwrap) {
// Strip error variants — success type is the first
// variant of the tagged return.
if (rt.kind != nkind.N_TTAGGED) { return nil; };
return rt.list;
};
// Plain call: declared return type is the local's type.
return rt;
};
// letslotsize — slot size for a `let` binding. Like slotsize, but
// detects `[_]T = arrlit;` (the type-AST has rhs == nil as the
// length-inferred sentinel) and computes count × element-size from
// the initialiser. Used by both scanlocals (prologue sizing) and
// cglet (slot alloc) so they agree on the frame layout.
//
// `let x = f();` (no annotation): infer from `f`'s declared return
// type so a 24B tagged-union return reserves all three spill slots,
// not the default 8B. Without this, the AX:DX:CX spill in cglet's
// tagged-init branch writes past the local and tramples the next
// slot.
export fn letslotsize(c: *cgen, n: *node) i32 = {
// `[_]T = arrlit;` — inferred-length array. slotsize would
// return elem_size * 1 (treating missing length as 1); intercept
@@ -727,7 +802,13 @@ export fn letslotsize(c: *cgen, n: *node) i32 = {
};
};
};
return slotsize(c, n.lhs);
if (n.lhs != nil) { return slotsize(c, n.lhs); };
// Annotation-less init: defer to the call's return type if we
// can infer it. Tagged-union returns need 24B; everything else
// matches slotsize on the inferred type.
let inferred: *node = inferletcalltype(c, n.rhs);
if (inferred != nil) { return slotsize(c, inferred); };
return 8;
};
fn slotsize(c: *cgen, typn: *node) i32 = {
@@ -779,6 +860,19 @@ fn slotsize(c: *cgen, typn: *node) i32 = {
// Named struct lookup.
let si: *structinfo = structlookup(c, nm);
if (si != nil) { return si.totsize; };
// Type alias (`type foo = !str;` / `type foo = bar;`):
// follow it so a tagged-union variant of a !str-aliased
// error type contributes 16 bytes to the max payload
// rather than 8 (the default).
if (c != nil) {
let aliased: *node = aliaslookup(c, nm);
if (aliased != nil) {
if (aliased.kind == nkind.N_TBANG) {
return slotsize(c, aliased.lhs);
};
return slotsize(c, aliased);
};
};
return 8;
};
if (k == nkind.N_TARRAY) {
@@ -930,7 +1024,20 @@ fn isstrtype(c: *cgen, t: *node) bool = {
if (isstrtyperaw(t)) { return true; };
if (c == nil) { return false; };
let r: *node = resolvetype(c, t);
return isstrtyperaw(r);
if (isstrtyperaw(r)) { return true; };
// `parserr = !str` — `!T` aliases shouldn't hide their
// underlying type from str-routing. Unwrap and re-check.
if (r != nil) {
if (r.kind == nkind.N_TBANG) {
let inner: *node = r.lhs;
if (isstrtyperaw(inner)) { return true; };
if (inner != nil) {
let r2: *node = resolvetype(c, inner);
if (isstrtyperaw(r2)) { return true; };
};
};
};
return false;
};
fn isslicetyperaw(t: *node) bool = {
@@ -1061,6 +1168,48 @@ export fn exprfloatkind(c: *cgen, n: *node) i32 = {
};
return 0;
};
if (k == nkind.N_DOT) {
// `p.field` where the struct field is f64/f32. Without this,
// `v.fval: i64` lowers to CVTSI on an integer-load value
// instead of CVTTSD2SI on the X0 the cgdot path actually
// emits for an f64 field.
let base: *node = n.lhs;
let fld: str = n.str;
if (base != nil) {
let sname: str;
sname.ptr = nil; sname.len = 0;
if (base.kind == nkind.N_IDENT) {
let lc: *local = localfindnode(c, base.str);
if (lc != nil) {
let tn: *node = lc.tnode;
if (tn != nil) {
if (tn.kind == nkind.N_TNAME) { sname = tn.str; };
if (tn.kind == nkind.N_TPTR) {
let pe: *node = tn.lhs;
if (pe != nil) {
if (pe.kind == nkind.N_TNAME) { sname = pe.str; };
};
};
};
};
};
if (sname.len > 0) {
let si: *structinfo = structlookup(c, sname);
if (si != nil) {
let fi: *fieldinfo = si.fields;
for (fi != nil) {
if (streq(fi.fname, fld)) {
if (isf32type(c, fi.tnode)) { return 1; };
if (isfloattype(c, fi.tnode)) { return 2; };
return 0;
};
fi = fi.finext;
};
};
};
};
return 0;
};
return 0;
};
@@ -1129,6 +1278,19 @@ fn rhstargetname(c: *cgen, rhs: *node) str = {
return nm;
};
if (rhs.kind == nkind.N_STRLIT) { return "str"; };
// `T{}` carries its type name on the lhs N_IDENT — the parser
// builds `N_STRUCTLIT{ lhs = N_IDENT("T"), list = fields }`.
// Needed so `return eof{};` (variant of a tagged union) resolves
// to the `eof` variant index rather than falling through to the
// "first non-str variant" fallback in taggedvariantindex.
if (rhs.kind == nkind.N_STRUCTLIT) {
let tref: *node = rhs.lhs;
if (tref != nil) {
if (tref.kind == nkind.N_IDENT) { return tref.str; };
if (tref.kind == nkind.N_TNAME) { return tref.str; };
};
return nm;
};
if (rhs.kind == nkind.N_IDENT) {
let lc: *local = localfindnode(c, rhs.str);
if (lc != nil) {
@@ -1154,7 +1316,7 @@ fn taggedvariantindex(c: *cgen, tagged: *node, rhs: *node) i32 = {
let idx: i32 = 0;
for (v != nil) {
if (v.kind == nkind.N_TNAME) {
if (streq(v.str, wantname)) { return idx; };
if (variantnamematch(v.str, wantname)) { return idx; };
};
v = v.next;
idx += 1;