fix(ir): resolve named-const array dims (0083) + materialize literal slice args (0084)
Two silent-miscompile codegen fixes:
0083 — named-const array dimension. `TypeResolver.resolveCompound`'s array
arm resolved the dimension with `if int_literal ... else 0`, so a named const
(`N :: 16; [N]T`) hit the silent `else 0`: the array became 0-length / 0-byte
and element access ran out of bounds (garbage for scalars, bus error for
slice/pointer/struct elements). The arm now delegates the dimension to
`inner.resolveArrayLen` (symmetric with `inner.resolveInner` for the element).
The stateful `Lowering.resolveArrayLen` evaluates it as a compile-time integer
across the comptime-constant / generic-value / module-global const tables and
emits a diagnostic — no fabricated length — when it isn't one.
0084 — `.[...]` literal passed directly as a call arg. `lowerArrayLiteral`
always yields an aggregate array value; the array→slice conversion is the
caller's job. The local-bound var-decl path did it, but the call-arg coercion
path had no array→slice arm, so `classify([N]T, []T)` returned `.none` and the
raw array was passed where a slice was expected (callee read its {ptr,len}
header off the wrong bytes → 0 / garbage / segfault). `classify` now returns a
new `.array_to_slice` plan for same-element `[N]T → []T`, and `coerceToType`
emits the existing `array_to_slice` op — identical to the local-bound path.
Regressions (fail-before/pass-after demonstrated on the pre-fix compiler):
examples/0140-types-named-const-array-dim.sx (s64 + string + struct elems)
examples/0141-types-slice-literal-direct-call-arg.sx (string + []s64)
Gate: zig build, zig build test, bash tests/run_examples.sh (387 passed).
Issues 0083 and 0084 marked RESOLVED.
This commit is contained in:
32
examples/0140-types-named-const-array-dim.sx
Normal file
32
examples/0140-types-named-const-array-dim.sx
Normal file
@@ -0,0 +1,32 @@
|
|||||||
|
// A fixed array whose dimension is a module-global named constant
|
||||||
|
// (`N :: 16; [N]T`) has the same layout as a literal-dimension array
|
||||||
|
// (`[16]T`): correct length and element stride for scalar, slice/pointer
|
||||||
|
// (string), and struct element types.
|
||||||
|
// Regression (issue 0083): a named-const dim resolved to length 0, giving a
|
||||||
|
// 0-byte alloca — scalar reads returned garbage and string/struct elements
|
||||||
|
// bus-errored.
|
||||||
|
#import "modules/std.sx";
|
||||||
|
|
||||||
|
N :: 4;
|
||||||
|
|
||||||
|
P :: struct { x: s64; y: s64; }
|
||||||
|
|
||||||
|
main :: () {
|
||||||
|
// Scalar elements: store then read back.
|
||||||
|
a : [N]s64 = ---;
|
||||||
|
a[0] = 7;
|
||||||
|
a[3] = 42;
|
||||||
|
print("scalar a0={} a3={}\n", a[0], a[3]);
|
||||||
|
|
||||||
|
// Slice/pointer elements (string): used to bus-error.
|
||||||
|
s : [N]string = ---;
|
||||||
|
s[0] = "hi";
|
||||||
|
s[1] = "yo";
|
||||||
|
print("string s0={} s1={}\n", s[0], s[1]);
|
||||||
|
|
||||||
|
// Struct elements.
|
||||||
|
ps : [N]P = ---;
|
||||||
|
ps[0] = P.{ x = 1, y = 2 };
|
||||||
|
ps[2] = P.{ x = 5, y = 6 };
|
||||||
|
print("struct p0x={} p0y={} p2x={}\n", ps[0].x, ps[0].y, ps[2].x);
|
||||||
|
}
|
||||||
34
examples/0141-types-slice-literal-direct-call-arg.sx
Normal file
34
examples/0141-types-slice-literal-direct-call-arg.sx
Normal file
@@ -0,0 +1,34 @@
|
|||||||
|
// A `.[...]` array/slice literal passed DIRECTLY as a call argument behaves
|
||||||
|
// identically to binding it to a typed local first: the literal is
|
||||||
|
// materialized into addressable storage and a {ptr,len} slice header is built
|
||||||
|
// over it, so the callee reads the element CONTENTS correctly.
|
||||||
|
// Regression (issue 0084): a direct literal arg passed the raw array value
|
||||||
|
// where a slice was expected, so the callee read its header off the wrong
|
||||||
|
// bytes and returned garbage (0).
|
||||||
|
#import "modules/std.sx";
|
||||||
|
|
||||||
|
count_nope :: (xs: []string) -> s64 {
|
||||||
|
n := 0;
|
||||||
|
i := 0;
|
||||||
|
while i < xs.len { if xs[i] == "nope" { n += 1; } i += 1; }
|
||||||
|
return n;
|
||||||
|
}
|
||||||
|
|
||||||
|
sum :: (xs: []s64) -> s64 {
|
||||||
|
s := 0;
|
||||||
|
i := 0;
|
||||||
|
while i < xs.len { s += xs[i]; i += 1; }
|
||||||
|
return s;
|
||||||
|
}
|
||||||
|
|
||||||
|
main :: () {
|
||||||
|
// string slice: direct literal vs local-bound — both see 2 "nope"s.
|
||||||
|
print("str direct={}\n", count_nope(.["a", "nope", "b", "nope"]));
|
||||||
|
local : []string = .["a", "nope", "b", "nope"];
|
||||||
|
print("str local={}\n", count_nope(local));
|
||||||
|
|
||||||
|
// numeric slice: direct literal vs local-bound — both sum to 100.
|
||||||
|
print("num direct={}\n", sum(.[10, 20, 30, 40]));
|
||||||
|
nums : []s64 = .[10, 20, 30, 40];
|
||||||
|
print("num local={}\n", sum(nums));
|
||||||
|
}
|
||||||
1
examples/expected/0140-types-named-const-array-dim.exit
Normal file
1
examples/expected/0140-types-named-const-array-dim.exit
Normal file
@@ -0,0 +1 @@
|
|||||||
|
0
|
||||||
@@ -0,0 +1 @@
|
|||||||
|
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
scalar a0=7 a3=42
|
||||||
|
string s0=hi s1=yo
|
||||||
|
struct p0x=1 p0y=2 p2x=5
|
||||||
@@ -0,0 +1 @@
|
|||||||
|
0
|
||||||
@@ -0,0 +1 @@
|
|||||||
|
|
||||||
@@ -0,0 +1,4 @@
|
|||||||
|
str direct=2
|
||||||
|
str local=2
|
||||||
|
num direct=100
|
||||||
|
num local=100
|
||||||
42
issues/0083-named-const-array-dimension-miscompiled.md
Normal file
42
issues/0083-named-const-array-dimension-miscompiled.md
Normal file
@@ -0,0 +1,42 @@
|
|||||||
|
# 0083 — fixed array with a named-constant dimension is miscompiled
|
||||||
|
|
||||||
|
> **RESOLVED.** Root cause: `TypeResolver.resolveCompound`'s array arm resolved
|
||||||
|
> the dimension with `if (length.data == .int_literal) ... else 0` — a named
|
||||||
|
> const (`N :: 16`) hit the silent `else 0`, so `[N]T` became a 0-length / 0-byte
|
||||||
|
> array and element access ran out of bounds (garbage for scalars, bus error for
|
||||||
|
> slice/pointer/struct elements). Fix: the array arm now delegates the dimension
|
||||||
|
> to `inner.resolveArrayLen` (symmetric with `inner.resolveInner` for the element
|
||||||
|
> type). The stateful `Lowering.resolveArrayLen` evaluates the dimension as a
|
||||||
|
> compile-time integer across the comptime-constant, generic-value, and
|
||||||
|
> module-global const tables, and emits a diagnostic (no fabricated length) when
|
||||||
|
> it isn't one. Files: `src/ir/type_resolver.zig`, `src/ir/lower.zig`,
|
||||||
|
> `src/ir/type_bridge.zig`. Regression: `examples/0140-types-named-const-array-dim.sx`
|
||||||
|
> (s64 + string + struct element types).
|
||||||
|
|
||||||
|
## Symptom
|
||||||
|
A fixed array whose dimension is a module-global integer constant (`N :: 16;
|
||||||
|
a : [N]T`) miscompiles element access: reads/writes compute a wrong address.
|
||||||
|
With `s64` elements `a[0]` returns GARBAGE (silent); with slice/pointer element
|
||||||
|
types (`[N]string`) it Bus-errors. The identical program with a LITERAL dimension
|
||||||
|
(`a : [16]T`) is correct. Silent-miscompile class (cf. 0079–0082).
|
||||||
|
|
||||||
|
## Reproduction
|
||||||
|
```sx
|
||||||
|
#import "modules/std.sx";
|
||||||
|
N :: 16;
|
||||||
|
main :: () { a : [N]s64 = ---; a[0] = 7; print("a0={}\n", a[0]); }
|
||||||
|
```
|
||||||
|
`./zig-out/bin/sx run` prints `a0=8472789232` (garbage); want `a0=7`. Replacing
|
||||||
|
`[N]` with `[16]` prints `7`.
|
||||||
|
|
||||||
|
## Investigation prompt
|
||||||
|
A fixed-array TYPE whose dimension is a named const (`N :: 16; [N]T`) resolves to
|
||||||
|
a wrong element stride / array length in codegen — element address computation is
|
||||||
|
wrong (garbage for scalars, bad pointer for slice/pointer elements). Literal
|
||||||
|
dimensions are correct, so the defect is in resolving the array-type DIMENSION
|
||||||
|
from a constant expression (vs a literal) — the dim likely resolves to 0/unknown
|
||||||
|
or the element size is wrong. Look at array-type resolution where the length is a
|
||||||
|
const-expr (type lowering / sizeof / element-stride computation). Fix so a
|
||||||
|
named-const dimension yields the same layout as the literal. Verify with the
|
||||||
|
repro (expect 7) + a `[N]string`/`[N]struct` case (no bus error, correct reads),
|
||||||
|
and `zig build && zig build test && bash tests/run_examples.sh` green.
|
||||||
43
issues/0084-slice-literal-direct-call-arg-miscompiled.md
Normal file
43
issues/0084-slice-literal-direct-call-arg-miscompiled.md
Normal file
@@ -0,0 +1,43 @@
|
|||||||
|
# 0084 — array/slice literal passed directly as a call argument miscompiles
|
||||||
|
|
||||||
|
> **RESOLVED.** Root cause: `lowerArrayLiteral` always produces an aggregate
|
||||||
|
> ARRAY value; the array→slice conversion is the caller's job. The local-bound
|
||||||
|
> var-decl path did it (emits `array_to_slice`), but the call-argument coercion
|
||||||
|
> path (`coerceCallArgs` → `coerceToType` → `CoercionResolver.classify`) had no
|
||||||
|
> array→slice arm, so `classify([N]T, []T)` returned `.none` and the raw array
|
||||||
|
> value was passed where a slice was expected — the callee read its {ptr,len}
|
||||||
|
> header off the wrong bytes (returned 0 / garbage, segfaulted for `[]s64`). Fix:
|
||||||
|
> `classify` now returns a new `.array_to_slice` plan for `[N]T → []T` (same
|
||||||
|
> element type), and `coerceToType` emits the existing `array_to_slice` op, which
|
||||||
|
> materializes the array into addressable storage and builds the slice header —
|
||||||
|
> identical to the local-bound path. Files: `src/ir/conversions.zig`,
|
||||||
|
> `src/ir/lower.zig`. Regression: `examples/0141-types-slice-literal-direct-call-arg.sx`
|
||||||
|
> (string + numeric `[]s64`, direct vs local-bound).
|
||||||
|
|
||||||
|
## Symptom
|
||||||
|
A `.[...]` array/slice literal passed DIRECTLY as a call argument yields a slice
|
||||||
|
whose element CONTENTS are not reliably readable in the callee (silent — reads
|
||||||
|
garbage, wrong results). Binding the same literal to a typed local first and
|
||||||
|
passing the local is correct.
|
||||||
|
|
||||||
|
## Reproduction
|
||||||
|
```sx
|
||||||
|
#import "modules/std.sx";
|
||||||
|
show :: (xs: []string) -> s64 { n:=0; i:=0; while i<xs.len { if xs[i]=="nope" {n+=1;} i+=1; } return n; }
|
||||||
|
main :: () {
|
||||||
|
print("direct={}\n", show(.["a","nope","b","nope"])); // prints 0 (WRONG)
|
||||||
|
local : []string = .["a","nope","b","nope"]; print("local={}\n", show(local)); // prints 2 (correct)
|
||||||
|
}
|
||||||
|
```
|
||||||
|
Want both `2`. Direct-literal-arg returns `0`.
|
||||||
|
|
||||||
|
## Investigation prompt
|
||||||
|
Passing a `.[...]` literal directly as a call arg builds a slice/array temporary
|
||||||
|
whose backing storage is not correctly materialized/kept alive for the callee —
|
||||||
|
the slice header may point at a stack temp that is clobbered, or the elements are
|
||||||
|
not stored before the call. Binding to a typed local first works (the local's
|
||||||
|
storage backs the slice). Look at how a literal aggregate argument is lowered at a
|
||||||
|
call site (materialize the literal into addressable storage whose lifetime spans
|
||||||
|
the call, then pass a slice/pointer to it) vs the local-bound path. Fix so a
|
||||||
|
directly-passed literal arg behaves identically to a local-bound one. Verify with
|
||||||
|
the repro (both `2`) + a numeric `[]s64` case, gate green.
|
||||||
@@ -42,6 +42,7 @@ pub const CoercionResolver = struct {
|
|||||||
ptr_int_bitcast, // ptr ↔ int
|
ptr_int_bitcast, // ptr ↔ int
|
||||||
widen, // same kind, dst wider
|
widen, // same kind, dst wider
|
||||||
narrow, // same kind, dst narrower
|
narrow, // same kind, dst narrower
|
||||||
|
array_to_slice, // [N]T → []T (materialize backing storage + header)
|
||||||
none, // nothing applies — pass the value through
|
none, // nothing applies — pass the value through
|
||||||
};
|
};
|
||||||
|
|
||||||
@@ -65,6 +66,20 @@ pub const CoercionResolver = struct {
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
// Fixed array → slice of the same element: an aggregate array value
|
||||||
|
// (e.g. a `.[...]` literal passed directly as a call arg) needs to be
|
||||||
|
// materialized into addressable storage and wrapped in a {ptr,len}
|
||||||
|
// header. Without this the array value is passed where a slice is
|
||||||
|
// expected — the callee reads the header off the wrong bytes (issue
|
||||||
|
// 0084). The local-bound path already does this conversion on its own.
|
||||||
|
if (!src_ty.isBuiltin() and !dst_ty.isBuiltin()) {
|
||||||
|
const si = self.l.module.types.get(src_ty);
|
||||||
|
const di = self.l.module.types.get(dst_ty);
|
||||||
|
if (si == .array and di == .slice and si.array.element == di.slice.element) {
|
||||||
|
return .array_to_slice;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
// Optional → Concrete unwrap (narrowing).
|
// Optional → Concrete unwrap (narrowing).
|
||||||
if (!src_ty.isBuiltin()) {
|
if (!src_ty.isBuiltin()) {
|
||||||
const src_info = self.l.module.types.get(src_ty);
|
const src_info = self.l.module.types.get(src_ty);
|
||||||
|
|||||||
@@ -11612,6 +11612,55 @@ pub const Lowering = struct {
|
|||||||
return self.resolveTypeWithBindings(node);
|
return self.resolveTypeWithBindings(node);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
/// Fixed-array dimension hook for `TypeResolver.resolveCompound`. A literal
|
||||||
|
/// `[16]T` and a named-const `N :: 16; [N]T` must resolve to the SAME length:
|
||||||
|
/// the dimension is a compile-time integer, looked up in the comptime / value
|
||||||
|
/// / module-const tables the stateful lowering owns. A dimension that isn't a
|
||||||
|
/// compile-time integer is a hard error — emitting a diagnostic (rather than
|
||||||
|
/// fabricating a 0 length, which gives a 0-byte array and out-of-bounds
|
||||||
|
/// element access, issue 0083).
|
||||||
|
pub fn resolveArrayLen(self: *Lowering, len_node: *const Node) u32 {
|
||||||
|
if (self.comptimeArrayDim(len_node)) |n| {
|
||||||
|
if (n < 0) {
|
||||||
|
if (self.diagnostics) |d|
|
||||||
|
d.addFmt(.err, len_node.span, "array dimension must be non-negative, got {}", .{n});
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
return @intCast(n);
|
||||||
|
}
|
||||||
|
if (self.diagnostics) |d|
|
||||||
|
d.addFmt(.err, len_node.span, "array dimension must be a compile-time integer constant", .{});
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
|
||||||
|
/// Evaluate a fixed-array dimension to a compile-time integer: a literal, or
|
||||||
|
/// a name bound to an integer in the comptime-constant (`OS`/loop cursors),
|
||||||
|
/// generic-value (`$N`), or module-global const (`N :: 16`) tables. Returns
|
||||||
|
/// null when the dimension isn't a compile-time integer.
|
||||||
|
fn comptimeArrayDim(self: *Lowering, node: *const Node) ?i64 {
|
||||||
|
return switch (node.data) {
|
||||||
|
.int_literal => |lit| lit.value,
|
||||||
|
.identifier => |id| self.comptimeIntNamed(id.name),
|
||||||
|
.type_expr => |te| self.comptimeIntNamed(te.name),
|
||||||
|
else => null,
|
||||||
|
};
|
||||||
|
}
|
||||||
|
|
||||||
|
/// Resolve a name to a compile-time integer across the three const tables.
|
||||||
|
fn comptimeIntNamed(self: *Lowering, name: []const u8) ?i64 {
|
||||||
|
if (self.comptime_constants.get(name)) |cv| switch (cv) {
|
||||||
|
.int_val => |iv| return iv,
|
||||||
|
else => {},
|
||||||
|
};
|
||||||
|
if (self.comptime_value_bindings) |cvb| {
|
||||||
|
if (cvb.get(name)) |v| return v;
|
||||||
|
}
|
||||||
|
if (self.program_index.module_const_map.get(name)) |ci| {
|
||||||
|
if (ci.value.data == .int_literal) return ci.value.data.int_literal.value;
|
||||||
|
}
|
||||||
|
return null;
|
||||||
|
}
|
||||||
|
|
||||||
/// Resolve a type node, checking type_bindings first for generic type params.
|
/// Resolve a type node, checking type_bindings first for generic type params.
|
||||||
pub fn resolveTypeWithBindings(self: *Lowering, node: *const Node) TypeId {
|
pub fn resolveTypeWithBindings(self: *Lowering, node: *const Node) TypeId {
|
||||||
// Pack-index in a type position: `$<pack>[<lit>]` resolves to the
|
// Pack-index in a type position: `$<pack>[<lit>]` resolves to the
|
||||||
@@ -13984,6 +14033,7 @@ pub const Lowering = struct {
|
|||||||
.ptr_int_bitcast => return self.builder.emit(.{ .bitcast = .{ .operand = val, .from = src_ty, .to = dst_ty } }, dst_ty),
|
.ptr_int_bitcast => return self.builder.emit(.{ .bitcast = .{ .operand = val, .from = src_ty, .to = dst_ty } }, dst_ty),
|
||||||
.narrow => return self.builder.emit(.{ .narrow = .{ .operand = val, .from = src_ty, .to = dst_ty } }, dst_ty),
|
.narrow => return self.builder.emit(.{ .narrow = .{ .operand = val, .from = src_ty, .to = dst_ty } }, dst_ty),
|
||||||
.widen => return self.builder.emit(.{ .widen = .{ .operand = val, .from = src_ty, .to = dst_ty } }, dst_ty),
|
.widen => return self.builder.emit(.{ .widen = .{ .operand = val, .from = src_ty, .to = dst_ty } }, dst_ty),
|
||||||
|
.array_to_slice => return self.builder.emit(.{ .array_to_slice = .{ .operand = val } }, dst_ty),
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|||||||
@@ -28,6 +28,18 @@ const StatelessInner = struct {
|
|||||||
pub fn resolveInner(self: StatelessInner, node: *const Node) TypeId {
|
pub fn resolveInner(self: StatelessInner, node: *const Node) TypeId {
|
||||||
return resolveAstType(node, self.table, self.alias_map);
|
return resolveAstType(node, self.table, self.alias_map);
|
||||||
}
|
}
|
||||||
|
/// Fixed-array dimension at registration time (no bindings / const tables).
|
||||||
|
/// Only a literal dimension is knowable here; a named-const dimension
|
||||||
|
/// (`N :: 16; [N]T`) is resolved by the stateful caller
|
||||||
|
/// (`Lowering.resolveArrayLen`) before it ever reaches this binding-free
|
||||||
|
/// path — mirroring how `pack_index_type_expr` is handled stateful-first.
|
||||||
|
pub fn resolveArrayLen(self: StatelessInner, len_node: *const Node) u32 {
|
||||||
|
_ = self;
|
||||||
|
return switch (len_node.data) {
|
||||||
|
.int_literal => |lit| @intCast(lit.value),
|
||||||
|
else => 0,
|
||||||
|
};
|
||||||
|
}
|
||||||
};
|
};
|
||||||
|
|
||||||
// ── AST Node → TypeId ───────────────────────────────────────────────────
|
// ── AST Node → TypeId ───────────────────────────────────────────────────
|
||||||
|
|||||||
@@ -23,6 +23,12 @@ const PrimInner = struct {
|
|||||||
else => .unresolved,
|
else => .unresolved,
|
||||||
};
|
};
|
||||||
}
|
}
|
||||||
|
pub fn resolveArrayLen(_: PrimInner, len_node: *const Node) u32 {
|
||||||
|
return switch (len_node.data) {
|
||||||
|
.int_literal => |lit| @intCast(lit.value),
|
||||||
|
else => 0,
|
||||||
|
};
|
||||||
|
}
|
||||||
};
|
};
|
||||||
|
|
||||||
test "TypeResolver.resolvePrimitive maps builtin keywords, null otherwise" {
|
test "TypeResolver.resolvePrimitive maps builtin keywords, null otherwise" {
|
||||||
|
|||||||
@@ -93,7 +93,12 @@ pub const TypeResolver = struct {
|
|||||||
.optional_type_expr => |ot| table.optionalOf(inner.resolveInner(ot.inner_type)),
|
.optional_type_expr => |ot| table.optionalOf(inner.resolveInner(ot.inner_type)),
|
||||||
.array_type_expr => |at| blk: {
|
.array_type_expr => |at| blk: {
|
||||||
const elem = inner.resolveInner(at.element_type);
|
const elem = inner.resolveInner(at.element_type);
|
||||||
const len: u32 = if (at.length.data == .int_literal) @intCast(at.length.data.int_literal.value) else 0;
|
// The dimension is delegated to `inner` exactly like the element
|
||||||
|
// type: a literal `[16]T` and a named-const `N :: 16; [N]T` must
|
||||||
|
// produce the same length. The stateful resolver consults the
|
||||||
|
// const tables; the binding-free one handles literal dims (issue
|
||||||
|
// 0083 — a 0 here gives a 0-byte array and OOB element access).
|
||||||
|
const len = inner.resolveArrayLen(at.length);
|
||||||
break :blk table.arrayOf(elem, len);
|
break :blk table.arrayOf(elem, len);
|
||||||
},
|
},
|
||||||
.function_type_expr => |ft| blk: {
|
.function_type_expr => |ft| blk: {
|
||||||
|
|||||||
Reference in New Issue
Block a user