Hojun-Cho
922877309b
Cascades the four enum kinds through every signature and local that
holds one of their values, then removes the type_assignable /
unify_arith relaxation that previously let bare i32 mix with the
named enum types.
Signature updates:
- kwlookup() now returns `tkind` (not i32); tokname() takes `tkind`
- accepttok / expecttok / bprec / isassignop take `tkind`
- parsearglist's closekind is `tkind`
- newtype / prim take `tykind`; scopedefine takes `skind`
- newnode / nkname take `nkind`
Struct fields:
- tok.kind is `tkind`; parser.curkind is `tkind`
- node.kind is `nkind`; node.op is `tkind`
- tinfo.kind is `tykind`; sym.skind is `skind`
Locals holding kinds across lex/parse/check/cgen are now typed with
their enum, including sentinel patterns like `let lkind: nkind =
nkind.N_NONE; if (...) lkind = tn.kind;`.
The selfhost cgen had a load-width bug exposed by this: fieldsize()
fell back to 8 bytes for any TNAME that wasn't a struct or primitive.
For a tkind-typed field that gave `MOVQ (BX), AX` instead of `MOVL`,
diverging from the C cgen on tok.kind / parser.curkind / etc. Two
fixes:
- fieldsize now consults the enum registry and returns the storage
type's size (4 for `enum i32`)
- collectenums runs before collectstructs in cgfile so the registry
is populated when registerstruct asks for field sizes
All 22 tests stay green; 990/993/995 byte-identity probes pass with
the strict typing in place.
1018 lines
31 KiB
Plaintext
1018 lines
31 KiB
Plaintext
// selfhost/cmd/wcc/cgenutil.ww — split out of cgen.ww.
|
||
//
|
||
// General helpers used across cgenexpr / cgenstmt / cgendecl:
|
||
// - pushargsrev: per-call arg pushing
|
||
// - type predicates: isstr*/isslice*/istagged*/nodeis* families
|
||
// - field ops: fieldloadop, fieldstoreop
|
||
// - index helpers: indexbaseesz, dotinnerstructptr, elemsizeof
|
||
// - slot sizing: structlookup, primsize, slotsize, fieldsize,
|
||
// registerstruct, collectstructs
|
||
// - rhs helpers: rhstargetname, taggedvariantindex
|
||
//
|
||
// Bundler pulls this in transitively via cgen.ww; consumers don't
|
||
// need to `use cgenutil;` directly.
|
||
|
||
use os;
|
||
use mem;
|
||
use ast;
|
||
use tok;
|
||
use typ;
|
||
use sym;
|
||
use strconv;
|
||
|
||
// ---- expression cgen -------------------------------------------------
|
||
|
||
// pushargsrev — recursively walks the arg list, evaluates rightmost
|
||
// first, and pushes. str args take two slots (ptr in AX, len in BX);
|
||
// the order on the stack so a left-to-right pop into argregs lands
|
||
// (ptr, len) correctly is: PUSHQ BX (top), PUSHQ AX (above) — the
|
||
// pop sequence then yields AX, then BX.
|
||
fn pushargsrev(c: *cgen, arg: *node) i32 = {
|
||
if (arg == nil) { return 0; };
|
||
let rest: i32 = pushargsrev(c, arg.next);
|
||
// nkind.N_SLICE expression as arg: `buf[lo:hi]` builds a slice header
|
||
// on the stack matching C cgen's sequence — push base, push hi,
|
||
// compute lo, pop into BX/CX, derive len/ptr, push (cap, len, ptr).
|
||
if (arg.kind == nkind.N_SLICE) {
|
||
let base: *node = arg.lhs;
|
||
let lo: *node = arg.rhs;
|
||
let hi: *node = arg.cond;
|
||
let baselocal: *local = nil;
|
||
if (base != nil) {
|
||
if (base.kind == nkind.N_IDENT) {
|
||
let bn: str = base.str;
|
||
baselocal = localfindnode(c, bn);
|
||
};
|
||
};
|
||
// base address → push
|
||
if (baselocal != nil) {
|
||
let tn: *node = baselocal.tnode;
|
||
if (tn != nil) {
|
||
if (tn.kind == nkind.N_TARRAY) {
|
||
emitline("\tLEAQ\t");
|
||
emitoff(baselocal.off: i64);
|
||
emitline("(BP), AX\n");
|
||
} else {
|
||
emitline("\tMOVQ\t");
|
||
emitoff(baselocal.off: i64);
|
||
emitline("(BP), AX\n");
|
||
};
|
||
} else {
|
||
emitline("\tMOVQ\t");
|
||
emitoff(baselocal.off: i64);
|
||
emitline("(BP), AX\n");
|
||
};
|
||
} else {
|
||
cgexpr(c, base);
|
||
};
|
||
emitline("\tPUSHQ\tAX\n");
|
||
// hi (default base length) → push
|
||
if (hi != nil) {
|
||
cgexpr(c, hi);
|
||
} else { if (baselocal != nil) {
|
||
let tn: *node = baselocal.tnode;
|
||
if (tn != nil) {
|
||
if (tn.kind == nkind.N_TARRAY) {
|
||
let lenn: *node = tn.rhs;
|
||
if (lenn != nil) {
|
||
if (lenn.kind == nkind.N_INTLIT) {
|
||
emitline("\tMOVQ\t$");
|
||
emituint(lenn.uval);
|
||
emitline(", AX\n");
|
||
};
|
||
};
|
||
} else { if (tn.kind == nkind.N_TSLICE) {
|
||
emitline("\tMOVQ\t");
|
||
emitoff((baselocal.off + 8): i64);
|
||
emitline("(BP), AX\n");
|
||
} else { if (tn.kind == nkind.N_TNAME) {
|
||
if (streq(tn.str, "str")) {
|
||
emitline("\tMOVQ\t");
|
||
emitoff((baselocal.off + 8): i64);
|
||
emitline("(BP), AX\n");
|
||
};
|
||
};};};
|
||
};
|
||
} else {
|
||
emitline("\tMOVQ\t$0, AX\n");
|
||
};};
|
||
emitline("\tPUSHQ\tAX\n");
|
||
// lo (default 0) → AX
|
||
if (lo != nil) { cgexpr(c, lo); }
|
||
else { emitline("\tMOVQ\t$0, AX\n"); };
|
||
emitline("\tPOPQ\tBX\n"); // hi
|
||
emitline("\tPOPQ\tCX\n"); // base
|
||
emitline("\tMOVQ\tBX, DX\n"); // DX = hi
|
||
emitline("\tSUBQ\tAX, DX\n"); // DX = hi - lo = len
|
||
emitline("\tADDQ\tAX, CX\n"); // CX = base + lo = ptr
|
||
emitline("\tPUSHQ\tDX\n"); // cap
|
||
emitline("\tPUSHQ\tDX\n"); // len
|
||
emitline("\tPUSHQ\tCX\n"); // ptr (top)
|
||
return rest + 3;
|
||
};
|
||
// Slice/tagged ident args: emit per-register MOVQ+PUSHQ pairs in
|
||
// reverse order (cap/v1, len/v0, ptr/tag) so a left-to-right pop
|
||
// into argregs lands the canonical (ptr/tag, len/v0, cap/v1).
|
||
if (arg.kind == nkind.N_IDENT) {
|
||
let nm: str = arg.str;
|
||
let lc: *local = localfindnode(c, nm);
|
||
if (lc != nil) {
|
||
let off: i32 = lc.off;
|
||
if (isslicetype(c, lc.tnode) || istaggedtype(lc.tnode)) {
|
||
emitline("\tMOVQ\t");
|
||
emitoff((off + 16): i64);
|
||
emitline("(BP), AX\n");
|
||
emitline("\tPUSHQ\tAX\n");
|
||
emitline("\tMOVQ\t");
|
||
emitoff((off + 8): i64);
|
||
emitline("(BP), AX\n");
|
||
emitline("\tPUSHQ\tAX\n");
|
||
emitline("\tMOVQ\t");
|
||
emitoff(off: i64);
|
||
emitline("(BP), AX\n");
|
||
emitline("\tPUSHQ\tAX\n");
|
||
return rest + 3;
|
||
};
|
||
};
|
||
};
|
||
cgexpr(c, arg);
|
||
if (nodeisslice(c, arg)) {
|
||
emitline("\tPUSHQ\tCX\n");
|
||
emitline("\tPUSHQ\tBX\n");
|
||
emitline("\tPUSHQ\tAX\n");
|
||
return rest + 3;
|
||
};
|
||
if (nodeisstr(c, arg)) {
|
||
emitline("\tPUSHQ\tBX\n");
|
||
emitline("\tPUSHQ\tAX\n");
|
||
return rest + 2;
|
||
};
|
||
emitline("\tPUSHQ\tAX\n");
|
||
return rest + 1;
|
||
};
|
||
|
||
fn nodeisslice(c: *cgen, n: *node) bool = {
|
||
if (n == nil) { return false; };
|
||
let k: nkind = n.kind;
|
||
if (k == nkind.N_IDENT) {
|
||
let nm: str = n.str;
|
||
let lc: *local = localfindnode(c, nm);
|
||
if (lc != nil) { return isslicetype(c, lc.tnode); };
|
||
return false;
|
||
};
|
||
if (k == nkind.N_SLICE) { return true; };
|
||
return false;
|
||
};
|
||
|
||
// nodeisstr — best-effort surface check: does this expression
|
||
// evaluate to a str value? Used to drive the call-arg push convention
|
||
// (str args take two slots: ptr + len).
|
||
fn nodeisstr(c: *cgen, n: *node) bool = {
|
||
if (n == nil) { return false; };
|
||
let k: nkind = n.kind;
|
||
if (k == nkind.N_STRLIT) { return true; };
|
||
if (k == nkind.N_IDENT) {
|
||
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; };
|
||
};
|
||
};
|
||
};
|
||
return false;
|
||
};
|
||
if (k == nkind.N_CALL) {
|
||
let callee: *node = n.lhs;
|
||
if (callee != nil) {
|
||
if (callee.kind == nkind.N_IDENT) {
|
||
let cnm: str = callee.str;
|
||
let rt: *node = fnretlookup(c, cnm);
|
||
return isstrtype(c, rt);
|
||
};
|
||
};
|
||
return false;
|
||
};
|
||
if (k == nkind.N_DOT) {
|
||
let base: *node = n.lhs;
|
||
let fld: str = n.str;
|
||
// `<expr>.ptr` is *u8 not str; `<expr>.len` is i32 not str.
|
||
if (streq(fld, "ptr")) { return false; };
|
||
if (streq(fld, "len")) { return false; };
|
||
if (streq(fld, "cap")) { return false; };
|
||
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;
|
||
let lkind: nkind = nkind.N_NONE;
|
||
if (tn != nil) { lkind = tn.kind; };
|
||
if (lkind == nkind.N_TNAME) { sname = tn.str; };
|
||
if (lkind == nkind.N_TPTR) {
|
||
let inner: *node = tn.lhs;
|
||
if (inner != nil) {
|
||
if (inner.kind == nkind.N_TNAME) { sname = inner.str; };
|
||
};
|
||
};
|
||
};
|
||
};
|
||
// Chained dot (`p.foo.bar`): use dotinnerstructptr
|
||
// to resolve the inner chain to the *struct it lands
|
||
// on, then look up `fld` in that struct.
|
||
if (base.kind == nkind.N_DOT) {
|
||
let innert: *node = dotinnerstructptr(c, base);
|
||
if (innert != nil) {
|
||
if (innert.kind == nkind.N_TNAME) { sname = innert.str; };
|
||
};
|
||
};
|
||
if (sname.len > 0) {
|
||
let si: *structinfo = structlookup(c, sname);
|
||
if (si != nil) {
|
||
let fi: *fieldinfo = si.fields;
|
||
for (fi != nil) {
|
||
let fn_: str = fi.fname;
|
||
if (streq(fn_, fld)) {
|
||
return isstrtype(c, fi.tnode);
|
||
};
|
||
fi = fi.finext;
|
||
};
|
||
};
|
||
};
|
||
};
|
||
return false;
|
||
};
|
||
if (k == nkind.N_CAST) {
|
||
return isstrtype(c, n.rhs);
|
||
};
|
||
return false;
|
||
};
|
||
|
||
// typenameisunsigned — true for u8/u16/u32/u64/uint/uintptr.
|
||
fn typenameisunsigned(nm: str) bool = {
|
||
if (streq(nm, "u8")) { return true; };
|
||
if (streq(nm, "u16")) { return true; };
|
||
if (streq(nm, "u32")) { return true; };
|
||
if (streq(nm, "u64")) { return true; };
|
||
if (streq(nm, "uint")) { return true; };
|
||
if (streq(nm, "uintptr")) { return true; };
|
||
return false;
|
||
};
|
||
|
||
// typenodeisunsigned — recurse through TNAME / TPTR / TSLICE etc.
|
||
fn typenodeisunsigned(t: *node) bool = {
|
||
if (t == nil) { return false; };
|
||
if (t.kind == nkind.N_TNAME) { return typenameisunsigned(t.str); };
|
||
return false;
|
||
};
|
||
|
||
// typeis8byteprimitive — does this type take exactly one 8-byte
|
||
// slot (pointer / fn-ptr / 64-bit int / chan / scalar primitive
|
||
// padded up to 8) rather than a wider aggregate? Used by nkind.N_LET
|
||
// zero-init to mirror C cgen's "only zero if sz == 8 at the type
|
||
// level" rule. Strings (16), slices (24), tagged unions (>=16),
|
||
// tuples (16), structs (varies), arrays — all fall through to
|
||
// false here even when their *slot* rounds up to 8.
|
||
fn typeis8byteprimitive(c: *cgen, t: *node) bool = {
|
||
if (t == nil) { return false; };
|
||
let k: nkind = t.kind;
|
||
if (k == nkind.N_TPTR) { return true; };
|
||
if (k == nkind.N_TFN) { return true; };
|
||
if (k == nkind.N_TCHAN) { return true; };
|
||
if (k == nkind.N_TSLICE) { return false; };
|
||
if (k == nkind.N_TARRAY) { return false; };
|
||
if (k == nkind.N_TTUPLE) { return false; };
|
||
if (k == nkind.N_TTAGGED){ return false; };
|
||
if (k == nkind.N_TNAME) {
|
||
let nm: str = t.str;
|
||
if (streq(nm, "str")) { return false; };
|
||
// Struct alias: not a primitive even if the slot is 8B.
|
||
if (structlookup(c, nm) != nil) { return false; };
|
||
// Primitive (i8/u8/.../i64/u64/bool/rune/f32/f64/int/...).
|
||
// All of these get slot-padded to 8 and zero-init in C.
|
||
if (primsize(nm) > 0) { return true; };
|
||
return false;
|
||
};
|
||
return false;
|
||
};
|
||
|
||
// typenameissigned — true for i8/i16/i32/i64/int/rune.
|
||
fn typenameissigned(nm: str) bool = {
|
||
if (streq(nm, "i8")) { return true; };
|
||
if (streq(nm, "i16")) { return true; };
|
||
if (streq(nm, "i32")) { return true; };
|
||
if (streq(nm, "i64")) { return true; };
|
||
if (streq(nm, "int")) { return true; };
|
||
if (streq(nm, "rune")) { return true; };
|
||
return false;
|
||
};
|
||
|
||
// fieldloadop — pick the load instruction for a non-str struct
|
||
// field by its declared size + signedness. Mirrors the C cgen op
|
||
// dispatch (MOVZBQ for u8/bool/i8, MOVSXD for i32, MOVL for u32, MOVQ
|
||
// for 8-byte). f might be nil for fields outside our struct registry.
|
||
fn fieldloadop(f: *fieldinfo) str = {
|
||
if (f == nil) { return "MOVQ"; };
|
||
let sz: i32 = f.fsz;
|
||
if (sz == 1) { return "MOVZBQ"; };
|
||
if (sz == 4) {
|
||
let t: *node = f.tnode;
|
||
if (t != nil) {
|
||
if (t.kind == nkind.N_TNAME) {
|
||
if (typenameissigned(t.str)) { return "MOVSXD"; };
|
||
};
|
||
};
|
||
return "MOVL";
|
||
};
|
||
return "MOVQ";
|
||
};
|
||
|
||
// fieldstoreop — pick the store instruction for a non-str struct
|
||
// field by its declared size. MOVB for 1, MOVL for 4, MOVQ for 8.
|
||
fn fieldstoreop(f: *fieldinfo) str = {
|
||
if (f == nil) { return "MOVQ"; };
|
||
let sz: i32 = f.fsz;
|
||
if (sz == 1) { return "MOVB"; };
|
||
if (sz == 4) { return "MOVL"; };
|
||
return "MOVQ";
|
||
};
|
||
|
||
// indexbaseesz — element size for `arr[i]` where the base is a
|
||
// chained-dot pseudo-field `s.ptr` (s being str/*str/slice/*slice).
|
||
// For str the element is one byte; for `[]T` / `*[]T` we drill into
|
||
// the slice element type.
|
||
fn indexbaseesz(c: *cgen, base: *node) i32 = {
|
||
if (base == nil) { return 8; };
|
||
if (base.kind != nkind.N_DOT) { return 8; };
|
||
let fld: str = base.str;
|
||
let inner: *node = base.lhs;
|
||
if (inner == nil) { return 8; };
|
||
if (inner.kind != nkind.N_IDENT) { return 8; };
|
||
let nm: str = inner.str;
|
||
let lc: *local = localfindnode(c, nm);
|
||
if (lc == nil) { return 8; };
|
||
let tn: *node = lc.tnode;
|
||
if (tn == nil) { return 8; };
|
||
|
||
// `.ptr` pseudo-field on str/slice → element of the str/slice.
|
||
if (streq(fld, "ptr")) {
|
||
let innert: *node = tn;
|
||
if (tn.kind == nkind.N_TPTR) { innert = tn.lhs; };
|
||
if (innert == nil) { return 8; };
|
||
if (innert.kind == nkind.N_TNAME) {
|
||
if (streq(innert.str, "str")) { return 1; };
|
||
};
|
||
if (innert.kind == nkind.N_TSLICE) { return elemsizeof(innert); };
|
||
return 8;
|
||
};
|
||
|
||
// Generic struct field: if it's *T, element size is T's size.
|
||
let lkind: nkind = tn.kind;
|
||
let sname: str;
|
||
sname.ptr = nil; sname.len = 0;
|
||
if (lkind == nkind.N_TNAME) { sname = tn.str; };
|
||
if (lkind == nkind.N_TPTR) {
|
||
let pinner: *node = tn.lhs;
|
||
if (pinner != nil) {
|
||
if (pinner.kind == nkind.N_TNAME) { sname = pinner.str; };
|
||
};
|
||
};
|
||
if (sname.len == 0) { return 8; };
|
||
let si: *structinfo = structlookup(c, sname);
|
||
if (si == nil) { return 8; };
|
||
let fi: *fieldinfo = si.fields;
|
||
for (fi != nil) {
|
||
let fn_: str = fi.fname;
|
||
if (streq(fn_, fld)) {
|
||
let ft: *node = fi.tnode;
|
||
if (ft == nil) { return 8; };
|
||
if (ft.kind == nkind.N_TPTR) {
|
||
let elem: *node = ft.lhs;
|
||
if (elem != nil) {
|
||
if (elem.kind == nkind.N_TNAME) {
|
||
if (streq(elem.str, "str")) { return 16; };
|
||
let ps: i32 = primsize(elem.str);
|
||
if (ps > 0) { return ps; };
|
||
};
|
||
};
|
||
return 8;
|
||
};
|
||
if (ft.kind == nkind.N_TSLICE) { return elemsizeof(ft); };
|
||
// str-typed field: indexing yields one byte
|
||
// (`n.s[i]` where .s is str — matches C cgen's
|
||
// MOVZBQ for byte indexing).
|
||
if (ft.kind == nkind.N_TNAME) {
|
||
if (streq(ft.str, "str")) { return 1; };
|
||
};
|
||
return 8;
|
||
};
|
||
fi = fi.finext;
|
||
};
|
||
return 8;
|
||
};
|
||
|
||
// dotinnerstructptr — for an nkind.N_DOT whose lhs is a chain of dots
|
||
// or an nkind.N_IDENT, walk the chain and return the nkind.N_TNAME tnode of the
|
||
// struct that the chain dereferences to (i.e., for `r.sym` where
|
||
// .sym is *lsym, return nkind.N_TNAME("lsym")). Returns nil if the chain
|
||
// doesn't resolve to a *struct.
|
||
//
|
||
// Used by the chained-DOT cgen path so `r.sym.val` knows the outer
|
||
// is a field of `lsym`.
|
||
fn dotinnerstructptr(c: *cgen, n: *node) *node = {
|
||
if (n == nil) { return nil; };
|
||
if (n.kind != nkind.N_DOT) { return nil; };
|
||
let base: *node = n.lhs;
|
||
let fld: str = n.str;
|
||
if (base == nil) { return nil; };
|
||
|
||
// Resolve base's struct tnode.
|
||
let baset: *node = nil;
|
||
if (base.kind == nkind.N_IDENT) {
|
||
let lc: *local = localfindnode(c, base.str);
|
||
if (lc == nil) { return nil; };
|
||
let tn: *node = lc.tnode;
|
||
if (tn == nil) { return nil; };
|
||
// base could be either struct-by-value (nkind.N_TNAME) or *struct (nkind.N_TPTR).
|
||
if (tn.kind == nkind.N_TNAME) { baset = tn; };
|
||
if (tn.kind == nkind.N_TPTR) { baset = tn.lhs; };
|
||
} else { if (base.kind == nkind.N_DOT) {
|
||
baset = dotinnerstructptr(c, base);
|
||
};};
|
||
if (baset == nil) { return nil; };
|
||
if (baset.kind != nkind.N_TNAME) { return nil; };
|
||
|
||
// Look up the struct, find the field, return the field's *struct.
|
||
let si: *structinfo = structlookup(c, baset.str);
|
||
if (si == nil) { return nil; };
|
||
let fi: *fieldinfo = si.fields;
|
||
for (fi != nil) {
|
||
if (streq(fi.fname, fld)) {
|
||
let ft: *node = fi.tnode;
|
||
if (ft == nil) { return nil; };
|
||
if (ft.kind != nkind.N_TPTR) { return nil; };
|
||
let inner: *node = ft.lhs;
|
||
if (inner == nil) { return nil; };
|
||
if (inner.kind != nkind.N_TNAME) { return nil; };
|
||
return inner;
|
||
};
|
||
fi = fi.finext;
|
||
};
|
||
return nil;
|
||
};
|
||
|
||
// elemsizeof — given the type node of an indexable (`*T`, `[]T`,
|
||
// `[N]T`, `str`), return the byte size of one element (1 for u8/i8/
|
||
// bool/str-byte, 8 otherwise — same shape as C cgen's esz fallback).
|
||
fn elemsizeof(t: *node) i32 = {
|
||
if (t == nil) { return 1; };
|
||
let k: nkind = t.kind;
|
||
let elem: *node = nil;
|
||
if (k == nkind.N_TPTR) { elem = t.lhs; };
|
||
if (k == nkind.N_TSLICE) { elem = t.lhs; };
|
||
if (k == nkind.N_TARRAY) { elem = t.lhs; };
|
||
if (k == nkind.N_TNAME) {
|
||
let nm: str = t.str;
|
||
if (streq(nm, "str")) { return 1; };
|
||
// Indexing a primitive name (rare): element size = the prim.
|
||
let ps: i32 = primsize(nm);
|
||
if (ps > 0) { return ps; };
|
||
return 1;
|
||
};
|
||
if (elem == nil) { return 1; };
|
||
if (elem.kind == nkind.N_TNAME) {
|
||
let nm: str = elem.str;
|
||
// str element is 16B (ptr+len). primsize returns 0 for it.
|
||
if (streq(nm, "str")) { return 16; };
|
||
let ps: i32 = primsize(nm);
|
||
if (ps > 0) { return ps; };
|
||
};
|
||
return 8;
|
||
};
|
||
|
||
// nodeisunsigned — best-effort cgen-time inference from the AST. We
|
||
// don't have a typed AST yet, so we walk surface nodes:
|
||
// nkind.N_INTLIT — never marked unsigned (no tsuffix plumbing yet)
|
||
// nkind.N_IDENT — look up the local's declared type
|
||
// nkind.N_DOT — look up the field's declared type via struct reg
|
||
// nkind.N_BIN / nkind.N_UN — recurse: unsigned if either operand is unsigned
|
||
// nkind.N_CAST — use the cast target type
|
||
//
|
||
// Conservative: if we can't tell, return false (signed). The cost of
|
||
// being wrong here is byte-different asm vs C, not bad runtime.
|
||
fn nodeisunsigned(c: *cgen, n: *node) bool = {
|
||
if (n == nil) { return false; };
|
||
let k: nkind = n.kind;
|
||
if (k == nkind.N_IDENT) {
|
||
let nm: str = n.str;
|
||
let lc: *local = localfindnode(c, nm);
|
||
if (lc != nil) { return typenodeisunsigned(lc.tnode); };
|
||
return false;
|
||
};
|
||
if (k == nkind.N_DOT) {
|
||
let base: *node = n.lhs;
|
||
let fld: str = n.str;
|
||
if (base != nil) {
|
||
if (base.kind == nkind.N_IDENT) {
|
||
let bn: str = base.str;
|
||
let lc: *local = localfindnode(c, bn);
|
||
if (lc != nil) {
|
||
let tn: *node = lc.tnode;
|
||
let lkind: nkind = nkind.N_NONE;
|
||
if (tn != nil) { lkind = tn.kind; };
|
||
let sname: str;
|
||
sname.ptr = nil; sname.len = 0;
|
||
if (lkind == nkind.N_TPTR) {
|
||
let inner: *node = tn.lhs;
|
||
if (inner != nil) {
|
||
if (inner.kind == nkind.N_TNAME) { sname = inner.str; };
|
||
};
|
||
};
|
||
if (lkind == nkind.N_TNAME) { sname = tn.str; };
|
||
if (sname.len > 0) {
|
||
let si: *structinfo = structlookup(c, sname);
|
||
if (si != nil) {
|
||
let fi: *fieldinfo = si.fields;
|
||
for (fi != nil) {
|
||
let fn_: str = fi.fname;
|
||
if (streq(fn_, fld)) {
|
||
return typenodeisunsigned(fi.tnode);
|
||
};
|
||
fi = fi.finext;
|
||
};
|
||
};
|
||
};
|
||
};
|
||
};
|
||
};
|
||
return false;
|
||
};
|
||
if (k == nkind.N_CAST) { return typenodeisunsigned(n.rhs); };
|
||
if (k == nkind.N_BIN) {
|
||
if (nodeisunsigned(c, n.lhs)) { return true; };
|
||
return nodeisunsigned(c, n.rhs);
|
||
};
|
||
if (k == nkind.N_UN) { return nodeisunsigned(c, n.lhs); };
|
||
// nkind.N_INDEX: `p[i]` is unsigned iff p's element type is unsigned.
|
||
// Walks the base local's declared type and pulls the element
|
||
// out — *u8 → u8, [N]u32 → u32, []u64 → u64. Without this the
|
||
// compare-codegen for `p[i] >= 48u8` falls back to signed JGE
|
||
// instead of JAE, diverging from C w6c on byte indexing.
|
||
if (k == nkind.N_INDEX) {
|
||
let base: *node = n.lhs;
|
||
if (base != nil) {
|
||
if (base.kind == nkind.N_IDENT) {
|
||
let lc: *local = localfindnode(c, base.str);
|
||
if (lc != nil) {
|
||
let tn: *node = lc.tnode;
|
||
if (tn != nil) {
|
||
let elem: *node = nil;
|
||
if (tn.kind == nkind.N_TPTR) { elem = tn.lhs; };
|
||
if (tn.kind == nkind.N_TARRAY) { elem = tn.lhs; };
|
||
if (tn.kind == nkind.N_TSLICE) { elem = tn.lhs; };
|
||
if (elem != nil) {
|
||
return typenodeisunsigned(elem);
|
||
};
|
||
};
|
||
};
|
||
};
|
||
};
|
||
return false;
|
||
};
|
||
return false;
|
||
};
|
||
// ---- type-driven slot sizing ----------------------------------------
|
||
|
||
fn structlookup(c: *cgen, name: str) *structinfo = {
|
||
let s: *structinfo = c.structs;
|
||
for (s != nil) {
|
||
let sn: str = s.sname;
|
||
if (streq(sn, name)) { return s; };
|
||
s = s.sinext;
|
||
};
|
||
return nil;
|
||
};
|
||
|
||
// primsize — size in bytes of a primitive type name (or 0 if not
|
||
// recognised as a primitive — the caller falls back to other paths).
|
||
// fldnumidx — parse a tuple field name like "0" / "1" / "12" into an
|
||
// index, or -1 if not all-digits. Used by cgdot to dispatch
|
||
// `t.0` / `t.1` against an nkind.N_TTUPLE local without pulling in strconv.
|
||
fn fldnumidx(s: str) i32 = {
|
||
if (s.len == 0) { return -1; };
|
||
let r: i32 = 0;
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
let b: u8 = s[i];
|
||
if (b < 48u8) { return -1; };
|
||
if (b > 57u8) { return -1; };
|
||
r = r * 10 + ((b - 48u8): i32);
|
||
i += 1;
|
||
};
|
||
return r;
|
||
};
|
||
|
||
fn primsize(name: str) i32 = {
|
||
if (streq(name, "u8")) { return 1; };
|
||
if (streq(name, "i8")) { return 1; };
|
||
if (streq(name, "bool")) { return 1; };
|
||
if (streq(name, "u16")) { return 2; };
|
||
if (streq(name, "i16")) { return 2; };
|
||
if (streq(name, "u32")) { return 4; };
|
||
if (streq(name, "i32")) { return 4; };
|
||
if (streq(name, "f32")) { return 4; };
|
||
if (streq(name, "u64")) { return 8; };
|
||
if (streq(name, "i64")) { return 8; };
|
||
if (streq(name, "uint")) { return 8; };
|
||
if (streq(name, "int")) { return 8; };
|
||
if (streq(name, "uintptr")) { return 8; };
|
||
if (streq(name, "f64")) { return 8; };
|
||
if (streq(name, "rune")) { return 4; };
|
||
if (streq(name, "void")) { return 0; };
|
||
return 0;
|
||
};
|
||
|
||
// 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.
|
||
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
|
||
// and compute the real count first.
|
||
if (n.lhs != nil) {
|
||
if (n.lhs.kind == nkind.N_TARRAY) {
|
||
if (n.lhs.rhs == nil) {
|
||
if (n.rhs != nil) {
|
||
if (n.rhs.kind == nkind.N_ARRLIT) {
|
||
let elemn: *node = n.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 cnt: i32 = 0;
|
||
let e: *node = n.rhs.list;
|
||
for (e != nil) {
|
||
let adv: bool = true;
|
||
if (e.kind == nkind.N_FIELD) {
|
||
if (streq(e.str, "...")) {
|
||
e = nil;
|
||
adv = false;
|
||
};
|
||
};
|
||
if (adv) {
|
||
cnt += 1;
|
||
e = e.next;
|
||
};
|
||
};
|
||
return esz * cnt;
|
||
};
|
||
};
|
||
};
|
||
};
|
||
};
|
||
return slotsize(c, n.lhs);
|
||
};
|
||
|
||
fn slotsize(c: *cgen, typn: *node) i32 = {
|
||
if (typn == nil) { return 8; };
|
||
let k: nkind = typn.kind;
|
||
if (k == nkind.N_TPTR) { return 8; };
|
||
if (k == nkind.N_TFN) { return 8; };
|
||
if (k == nkind.N_TCHAN) { return 8; };
|
||
if (k == nkind.N_TSLICE) { return 24; };
|
||
if (k == nkind.N_TTUPLE) {
|
||
// Sum element sizes. Mirrors C cgen which uses raw type
|
||
// sizes; padding to 8 happens inside slotsize for primitives,
|
||
// so a `(i64, str)` resolves to 8 + 16 = 24 (matches the C
|
||
// cgen 24B init / positional-access layout).
|
||
let total: i32 = 0;
|
||
let p: *node = typn.list;
|
||
for (p != nil) {
|
||
total += slotsize(c, p);
|
||
p = p.next;
|
||
};
|
||
return total;
|
||
};
|
||
if (k == nkind.N_TTAGGED){
|
||
// Nullable `(*T | void)` collapses to a single 8B pointer.
|
||
if (isnullabletype(typn)) { return 8; };
|
||
// Slot = 8 (tag) + max(variant payload sizes), rounded up
|
||
// to an 8-byte multiple so the reg-passing ABI (size/8
|
||
// words) doesn't drop the last value register. Mirrors C
|
||
// cgen's resolve_type for nkind.N_TTAGGED.
|
||
let v: *node = typn.list;
|
||
let maxsz: i32 = 0;
|
||
for (v != nil) {
|
||
let sz: i32 = slotsize(c, v);
|
||
if (sz > maxsz) { maxsz = sz; };
|
||
v = v.next;
|
||
};
|
||
let pad: i32 = (maxsz + 7) & ~7;
|
||
return 8 + pad;
|
||
};
|
||
if (k == nkind.N_TNAME) {
|
||
let nm: str = typn.str;
|
||
if (streq(nm, "str")) { return 16; };
|
||
let ps: i32 = primsize(nm);
|
||
if (ps > 0) {
|
||
// Pad to 8 for stack slots — matches C cgen which spills
|
||
// every primitive into an 8-byte slot.
|
||
return 8;
|
||
};
|
||
// Named struct lookup.
|
||
let si: *structinfo = structlookup(c, nm);
|
||
if (si != nil) { return si.totsize; };
|
||
return 8;
|
||
};
|
||
if (k == nkind.N_TARRAY) {
|
||
let lenn: *node = typn.rhs;
|
||
let elemn: *node = typn.lhs;
|
||
let elen: i64 = 1i64;
|
||
if (lenn != nil) {
|
||
if (lenn.kind == nkind.N_INTLIT) { elen = lenn.uval: i64; };
|
||
};
|
||
let esz: i32 = 8;
|
||
if (elemn != nil) {
|
||
if (elemn.kind == nkind.N_TNAME) {
|
||
let en: str = elemn.str;
|
||
let ps: i32 = primsize(en);
|
||
if (ps > 0) { esz = ps; };
|
||
};
|
||
};
|
||
return (esz: i64 * elen): i32;
|
||
};
|
||
if (k == nkind.N_TSTRUCT) {
|
||
// Inline anonymous struct — sum of field sizes.
|
||
let f: *node = typn.list;
|
||
let total: i32 = 0;
|
||
for (f != nil) {
|
||
if (f.kind == nkind.N_TFIELD) {
|
||
total += slotsize(c, f.lhs);
|
||
};
|
||
f = f.next;
|
||
};
|
||
return total;
|
||
};
|
||
return 8;
|
||
};
|
||
|
||
// registerstruct — compute field offsets + total size for a struct
|
||
// type-decl, store in c.structs. Field type sizes use the same
|
||
// slotsize logic (with primitives kept at their natural width — we
|
||
// only round to 8 for stack slots, not struct interiors).
|
||
fn fieldsize(c: *cgen, tnode: *node) i32 = {
|
||
if (tnode == nil) { return 8; };
|
||
let k: nkind = tnode.kind;
|
||
if (k == nkind.N_TNAME) {
|
||
let nm: str = tnode.str;
|
||
if (streq(nm, "str")) { return 16; };
|
||
let ps: i32 = primsize(nm);
|
||
if (ps > 0) { return ps; };
|
||
let si: *structinfo = structlookup(c, nm);
|
||
if (si != nil) { return si.totsize; };
|
||
// Enum: size of its storage type. Mirrors the C cgen, which
|
||
// reads Type.size off the TY_ENUM (which inherits from .sub).
|
||
let en: *enumtype = enumlookup(c, nm);
|
||
if (en != nil) {
|
||
if (en.storage != nil) {
|
||
if (en.storage.kind == nkind.N_TNAME) {
|
||
let sps: i32 = primsize(en.storage.str);
|
||
if (sps > 0) { return sps; };
|
||
};
|
||
};
|
||
return 4; // default storage is i32
|
||
};
|
||
return 8;
|
||
};
|
||
if (k == nkind.N_TPTR) { return 8; };
|
||
if (k == nkind.N_TSLICE) { return 24; };
|
||
if (k == nkind.N_TARRAY) {
|
||
// Same shape as slotsize's TARRAY branch.
|
||
let lenn: *node = tnode.rhs;
|
||
let elemn: *node = tnode.lhs;
|
||
let elen: i64 = 1i64;
|
||
if (lenn != nil) {
|
||
if (lenn.kind == nkind.N_INTLIT) { elen = lenn.uval: i64; };
|
||
};
|
||
let esz: i32 = fieldsize(c, elemn);
|
||
return (esz: i64 * elen): i32;
|
||
};
|
||
return 8;
|
||
};
|
||
|
||
fn registerstruct(c: *cgen, name: str, tstruct: *node) void = {
|
||
let si: *structinfo = amalloc(c.a, 64u64): *structinfo;
|
||
si.sname = name;
|
||
si.fields = nil;
|
||
si.totsize = 0;
|
||
let head: *fieldinfo = nil;
|
||
let tail: *fieldinfo = nil;
|
||
let off: i32 = 0;
|
||
let f: *node = tstruct.list;
|
||
for (f != nil) {
|
||
if (f.kind == nkind.N_TFIELD) {
|
||
let sz: i32 = fieldsize(c, f.lhs);
|
||
// Align to 8 for any field >= 4 bytes (matches our other
|
||
// cgen choices). i8/u8/bool may sit on odd byte offsets;
|
||
// the C cgen does similar best-effort packing.
|
||
let aln: i32 = 1;
|
||
if (sz >= 8) { aln = 8; }
|
||
else { if (sz >= 4) { aln = 4; }
|
||
else { if (sz >= 2) { aln = 2; }; }; };
|
||
if ((off & (aln - 1)) != 0) {
|
||
off = (off + aln - 1) & ~(aln - 1);
|
||
};
|
||
let fi: *fieldinfo = amalloc(c.a, 48u64): *fieldinfo;
|
||
fi.fname = f.str;
|
||
fi.foff = off;
|
||
fi.fsz = sz;
|
||
fi.tnode = f.lhs;
|
||
if (head == nil) { head = fi; tail = fi; }
|
||
else { tail.finext = fi; tail = fi; };
|
||
off += sz;
|
||
};
|
||
f = f.next;
|
||
};
|
||
// Round total to 8 for stack-slot use.
|
||
if ((off & 7) != 0) { off = (off + 7) & ~7; };
|
||
si.fields = head;
|
||
si.totsize = off;
|
||
si.sinext = c.structs;
|
||
c.structs = si;
|
||
};
|
||
|
||
fn collectstructs(c: *cgen, file: *node) void = {
|
||
c.structs = nil;
|
||
if (file == nil) { return; };
|
||
let d: *node = file.list;
|
||
for (d != nil) {
|
||
if (d.kind == nkind.N_TYPEDECL) {
|
||
let body: *node = d.lhs;
|
||
if (body != nil) {
|
||
if (body.kind == nkind.N_TSTRUCT) {
|
||
registerstruct(c, d.str, body);
|
||
};
|
||
};
|
||
};
|
||
d = d.next;
|
||
};
|
||
};
|
||
|
||
// `type X = str;` aliases) to `str`. Takes *cgen so it can walk the
|
||
// alias chain registered at file load.
|
||
fn isstrtyperaw(t: *node) bool = {
|
||
if (t == nil) { return false; };
|
||
if (t.kind == nkind.N_TNAME) {
|
||
let nm: str = t.str;
|
||
if (streq(nm, "str")) { return true; };
|
||
};
|
||
return false;
|
||
};
|
||
|
||
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);
|
||
};
|
||
|
||
fn isslicetyperaw(t: *node) bool = {
|
||
if (t == nil) { return false; };
|
||
if (t.kind == nkind.N_TSLICE) { return true; };
|
||
return false;
|
||
};
|
||
|
||
fn isslicetype(c: *cgen, t: *node) bool = {
|
||
if (isslicetyperaw(t)) { return true; };
|
||
if (c == nil) { return false; };
|
||
let r: *node = resolvetype(c, t);
|
||
return isslicetyperaw(r);
|
||
};
|
||
|
||
fn istaggedtype(t: *node) bool = {
|
||
if (t == nil) { return false; };
|
||
if (t.kind == nkind.N_TTAGGED) { return true; };
|
||
return false;
|
||
};
|
||
|
||
// isnullabletype — nkind.N_TTAGGED with exactly two children, one *T and
|
||
// one `void`. Folds to a single 8-byte pointer slot per Hare's
|
||
// `(*T | null)` semantics. Mirrors check.c's resolve_type detection.
|
||
export fn isnullabletype(t: *node) bool = {
|
||
if (t == nil) { return false; };
|
||
if (t.kind != nkind.N_TTAGGED) { return false; };
|
||
let a: *node = t.list;
|
||
if (a == nil) { return false; };
|
||
let b: *node = a.next;
|
||
if (b == nil) { return false; };
|
||
if (b.next != nil) { return false; };
|
||
let aptr: bool = (a.kind == nkind.N_TPTR);
|
||
let bptr: bool = (b.kind == nkind.N_TPTR);
|
||
let avoid: bool = (a.kind == nkind.N_TNAME);
|
||
if (avoid) { avoid = streq(a.str, "void"); };
|
||
let bvoid: bool = (b.kind == nkind.N_TNAME);
|
||
if (bvoid) { bvoid = streq(b.str, "void"); };
|
||
if (aptr) { if (bvoid) { return true; }; };
|
||
if (avoid) { if (bptr) { return true; }; };
|
||
return false;
|
||
};
|
||
|
||
// nullableptrtag — 0-based index of the *T variant in a nullable
|
||
// union. The void variant takes the other slot (0 or 1).
|
||
export fn nullableptrtag(t: *node) i32 = {
|
||
if (t == nil) { return 0; };
|
||
if (t.kind != nkind.N_TTAGGED) { return 0; };
|
||
let a: *node = t.list;
|
||
if (a != nil) { if (a.kind == nkind.N_TPTR) { return 0; }; };
|
||
return 1;
|
||
};
|
||
|
||
// voidvariantindex — find the 0-based index of the `void` variant in a
|
||
// tagged-union type expr, -1 if absent. Used by cgreturn to map bare
|
||
// `return;` in a tagged-union-returning fn to the void variant's tag.
|
||
fn voidvariantindex(tagged: *node) i32 = {
|
||
if (tagged == nil) { return -1; };
|
||
if (tagged.kind != nkind.N_TTAGGED) { return -1; };
|
||
let v: *node = tagged.list;
|
||
let idx: i32 = 0;
|
||
for (v != nil) {
|
||
if (v.kind == nkind.N_TNAME) {
|
||
if (streq(v.str, "void")) { return idx; };
|
||
};
|
||
v = v.next;
|
||
idx += 1;
|
||
};
|
||
return -1;
|
||
};
|
||
|
||
// rhstargetname — for a returned value, what's its declared (or
|
||
// surface-inferred) type name? `expr: T` casts dictate T directly;
|
||
// bare strlit/intlit fall back to a primitive name.
|
||
fn rhstargetname(c: *cgen, rhs: *node) str = {
|
||
let nm: str;
|
||
nm.ptr = nil; nm.len = 0;
|
||
if (rhs == nil) { return nm; };
|
||
if (rhs.kind == nkind.N_CAST) {
|
||
let t: *node = rhs.rhs;
|
||
if (t != nil) {
|
||
if (t.kind == nkind.N_TNAME) { return t.str; };
|
||
};
|
||
return nm;
|
||
};
|
||
if (rhs.kind == nkind.N_STRLIT) { return "str"; };
|
||
if (rhs.kind == nkind.N_IDENT) {
|
||
let lc: *local = localfindnode(c, rhs.str);
|
||
if (lc != nil) {
|
||
let tn: *node = lc.tnode;
|
||
if (tn != nil) {
|
||
if (tn.kind == nkind.N_TNAME) { return tn.str; };
|
||
};
|
||
};
|
||
};
|
||
return nm;
|
||
};
|
||
|
||
// taggedvariantindex — given the tagged-union type expr and the
|
||
// returned value's surface type, find the matching variant's 0-based
|
||
// index. Compare by exact type name first; if no match, fall back to
|
||
// "any str-shape variant matches an str-typed value".
|
||
fn taggedvariantindex(c: *cgen, tagged: *node, rhs: *node) i32 = {
|
||
if (tagged == nil) { return -1; };
|
||
if (rhs == nil) { return -1; };
|
||
let wantname: str = rhstargetname(c, rhs);
|
||
if (wantname.len > 0) {
|
||
let v: *node = tagged.list;
|
||
let idx: i32 = 0;
|
||
for (v != nil) {
|
||
if (v.kind == nkind.N_TNAME) {
|
||
if (streq(v.str, wantname)) { return idx; };
|
||
};
|
||
v = v.next;
|
||
idx += 1;
|
||
};
|
||
};
|
||
// Fallback: by str-shape (resolves aliases).
|
||
let wantstr: bool = nodeisstr(c, rhs);
|
||
let v: *node = tagged.list;
|
||
let idx: i32 = 0;
|
||
for (v != nil) {
|
||
let visstr: bool = false;
|
||
if (v.kind == nkind.N_TNAME) {
|
||
if (isstrtype(c, v)) { visstr = true; };
|
||
};
|
||
if (visstr == wantstr) { return idx; };
|
||
v = v.next;
|
||
idx += 1;
|
||
};
|
||
return -1;
|
||
};
|