aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/CodeGen.cc
diff options
context:
space:
mode:
author_ov4 <ov4@ov4.org>2026-07-26 18:10:10 +0000
committer_ov4 <ov4@ov4.org>2026-07-26 18:10:10 +0000
commitc2c2e1771bf9b747f38dc145b5201b45c579d465 (patch)
tree09c752f360e56f523eb87e4a6dc17751be56210b /lib/CodeGen/CodeGen.cc
parenta44f1246dbe24fc813a212a5ef3f5811b1d6402a (diff)
fix: close type-safety holes found by adversarial review; update docs
Review findings (all reproduced, then fixed and re-verified): - '^x' leaked as a raw address into any builtin context (s <- ^x printed the variable's bytes as a C string; NOT ^x, '&', and ^x = ^x compiled). Address-of is now rejected everywhere except a pointer-typed target. - A whole-array variable passed the nominal gate as a value of its ELEMENT type (s <- arrayOfSeason stored the raw data pointer; = on it emitted invalid IR). Whole arrays are now rejected as operands and as coercion sources; array parameters are unaffected. - Enum values smuggled into INTEGER through array subscripts and DECLARE bounds (a[Autumn], ARRAY[Spring:Winter]); both now go through emitCoercedExpr like every other INTEGER context. - An array DECLAREd inside an IF/WHILE/REPEAT/FOR body and used after the block was a wild-pointer write (constant bounds) or a verifier dump (runtime bounds). Arrays are now block-scoped: control-flow bodies snapshot and restore the ArrayTable, so escape is a clean compile error. - Duplicate parameter names silently shadowed (BYREF write-back defeated); now a parse error. Trailing comma in parameter lists accepted; now a parse error consistent with every other list. - A variable/array/parameter/function could reuse a TYPE name; reserved. - RETURN with a value inside a PROCEDURE was silently discarded (control did not return); now an error, as is a bare RETURN in a FUNCTION. - INPUT r.f mis-parsed into a three-error cascade; now one clear error. - Assigning to an enum constant now says so instead of 'Unknown variable'. Docs: ARCH/CONVENTIONS/EXTENDING/SUBSYSTEMS/PITFALLS rewritten for the TYPE system (ABI additions, nominal-typing invariants, admission-gate doors, array-parameter ABI, new recipes, new sharp edges); README features/TODO updated — random files are unblocked. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Diffstat (limited to 'lib/CodeGen/CodeGen.cc')
-rw-r--r--lib/CodeGen/CodeGen.cc50
1 files changed, 47 insertions, 3 deletions
diff --git a/lib/CodeGen/CodeGen.cc b/lib/CodeGen/CodeGen.cc
index 64e1b9f..0e372d0 100644
--- a/lib/CodeGen/CodeGen.cc
+++ b/lib/CodeGen/CodeGen.cc
@@ -402,6 +402,11 @@ void CodeGen::emitDeclareStmt(DeclareStmtAST *Stmt) {
Stmt->getName().c_str());
return;
}
+ if (Types->resolve(Stmt->getName())) {
+ reportError("'%s' is a type name and cannot be declared as a variable",
+ Stmt->getName().c_str());
+ return;
+ }
Function *TheFunction = Builder->GetInsertBlock()->getParent();
AllocaInst *Alloca = CreateEntryBlockAlloca(TheFunction, Info->LLVMType, Stmt->getName());
@@ -603,6 +608,18 @@ Value *CodeGen::emitExpr(ExprAST *Expr) {
}
if (auto *Bin = dynamic_cast<BinaryExprAST*>(Expr)) {
+ // Whole arrays evaluate to raw data pointers and ^x to raw addresses;
+ // neither is a value any operator may consume.
+ if (isWholeArrayVar(Bin->getLHS()) || isWholeArrayVar(Bin->getRHS())) {
+ reportError("A whole array cannot be used as an operand; index it");
+ return nullptr;
+ }
+ if (dynamic_cast<AddrOfExprAST*>(Bin->getLHS()) ||
+ dynamic_cast<AddrOfExprAST*>(Bin->getRHS())) {
+ reportError("'^' (address-of) can only be used where a pointer type is expected");
+ return nullptr;
+ }
+
// Enum/Record/Pointer operands never reach ArithmeticHandler or the
// '&' path: their admissible operations are decided by name here.
const TypeInfo *LT = getExprTypeInfo(Bin->getLHS());
@@ -825,13 +842,18 @@ void CodeGen::emitIfStmt(IfStmtAST *Stmt) {
Builder->CreateCondBr(CondV, ThenBB, ElseBB);
+ // Arrays are block-scoped: metadata SSA values and the buffer setup live
+ // inside the branch, so a declaration must not survive it.
+ auto SavedArrays = Arrays->snapshotTable();
Builder->SetInsertPoint(ThenBB);
for (const auto &S : Stmt->getThenStmts()) emitStmt(S.get());
if (!Builder->GetInsertBlock()->getTerminator()) Builder->CreateBr(MergeBB);
+ Arrays->exchangeTable(SavedArrays);
Builder->SetInsertPoint(ElseBB);
for (const auto &S : Stmt->getElseStmts()) emitStmt(S.get());
if (!Builder->GetInsertBlock()->getTerminator()) Builder->CreateBr(MergeBB);
+ Arrays->exchangeTable(std::move(SavedArrays));
Builder->SetInsertPoint(MergeBB);
}
@@ -851,8 +873,11 @@ void CodeGen::emitWhileStmt(WhileStmtAST *Stmt) {
Builder->CreateCondBr(CondV, LoopBB, AfterBB);
+ // Arrays declared in the body are block-scoped (see emitIfStmt).
+ auto SavedArrays = Arrays->snapshotTable();
Builder->SetInsertPoint(LoopBB);
for (const auto &S : Stmt->getBody()) emitStmt(S.get());
+ Arrays->exchangeTable(std::move(SavedArrays));
if (!Builder->GetInsertBlock()->getTerminator())
Builder->CreateBr(CondBB);
@@ -869,8 +894,11 @@ void CodeGen::emitRepeatStmt(RepeatStmtAST *Stmt) {
Builder->CreateBr(LoopBB);
+ // Arrays declared in the body are block-scoped (see emitIfStmt).
+ auto SavedArrays = Arrays->snapshotTable();
Builder->SetInsertPoint(LoopBB);
for (const auto &S : Stmt->getBody()) emitStmt(S.get());
+ Arrays->exchangeTable(std::move(SavedArrays));
if (!Builder->GetInsertBlock()->getTerminator())
Builder->CreateBr(CondBB);
@@ -934,8 +962,11 @@ void CodeGen::emitForStmt(ForStmtAST *Stmt) {
Builder->CreateCondBr(CondV, LoopBB, AfterBB);
+ // Arrays declared in the body are block-scoped (see emitIfStmt).
+ auto SavedArrays = Arrays->snapshotTable();
Builder->SetInsertPoint(LoopBB);
for (const auto &S : Stmt->getBody()) emitStmt(S.get());
+ Arrays->exchangeTable(std::move(SavedArrays));
if (!Builder->GetInsertBlock()->getTerminator())
Builder->CreateBr(IncBB);
@@ -984,10 +1015,19 @@ void CodeGen::emitStmt(StmtAST *Stmt) {
}
if (auto *Ret = dynamic_cast<ReturnStmtAST*>(Stmt)) {
+ Function *F = Builder->GetInsertBlock()->getParent();
+ const FuncSig *Sig = FuncGen->getSignature(F->getName().str());
+ if (Sig && Sig->ReturnTypeName == "VOID" && Ret->getRetVal()) {
+ reportError("RETURN with a value is only allowed in a FUNCTION");
+ return;
+ }
+ if (Sig && Sig->ReturnTypeName != "VOID" && !Ret->getRetVal()) {
+ reportError("RETURN in a FUNCTION requires a value");
+ return;
+ }
Value *RetVal = nullptr;
if (Ret->getRetVal()) {
- Function *F = Builder->GetInsertBlock()->getParent();
- if (const FuncSig *Sig = FuncGen->getSignature(F->getName().str())) {
+ if (Sig) {
RetVal = emitCoercedExpr(Ret->getRetVal(), resolveType(Sig->ReturnTypeName));
} else {
RetVal = emitExpr(Ret->getRetVal());
@@ -1006,7 +1046,11 @@ void CodeGen::emitStmt(StmtAST *Stmt) {
if (auto *Assign = dynamic_cast<AssignStmtAST*>(Stmt)) {
const SymbolInfo *Info = getSymbolInfo(Assign->getName());
if (!Info) {
- reportError("Unknown variable name %s", Assign->getName().c_str());
+ if (Types->lookupEnumConstant(Assign->getName())) {
+ reportError("Enum value %s is not an assignable location", Assign->getName().c_str());
+ } else {
+ reportError("Unknown variable name %s", Assign->getName().c_str());
+ }
return;
}
if (Info->IsArray) {