[clang] [analyzer] Model overflow builtins (PR #102602)
DonĂ¡t Nagy via cfe-commits
cfe-commits at lists.llvm.org
Mon Sep 2 06:21:23 PDT 2024
================
@@ -50,6 +118,75 @@ class BuiltinFunctionChecker : public Checker<eval::Call> {
} // namespace
+std::pair<bool, bool>
+BuiltinFunctionChecker::checkOverflow(CheckerContext &C, SVal RetVal,
+ QualType Res) const {
+ ProgramStateRef State = C.getState();
+ SValBuilder &SVB = C.getSValBuilder();
+ ASTContext &ACtx = C.getASTContext();
+
+ // Calling a builtin with a non-integer type result produces compiler error.
+ assert(Res->isIntegerType());
+
+ unsigned BitWidth = ACtx.getIntWidth(Res);
+ auto MinVal =
+ llvm::APSInt::getMinValue(BitWidth, Res->isUnsignedIntegerType());
+ auto MaxVal =
+ llvm::APSInt::getMaxValue(BitWidth, Res->isUnsignedIntegerType());
+
+ SVal IsLeMax =
+ SVB.evalBinOp(State, BO_LE, RetVal, nonloc::ConcreteInt(MaxVal), Res);
+ SVal IsGeMin =
+ SVB.evalBinOp(State, BO_GE, RetVal, nonloc::ConcreteInt(MinVal), Res);
+
+ auto [MayNotOverflow, MayOverflow] =
+ State->assume(IsLeMax.castAs<DefinedOrUnknownSVal>());
+ auto [MayNotUnderflow, MayUnderflow] =
+ State->assume(IsGeMin.castAs<DefinedOrUnknownSVal>());
+
+ return {MayOverflow || MayUnderflow, MayNotOverflow && MayNotUnderflow};
+}
+
+void BuiltinFunctionChecker::handleOverflowBuiltin(const CallEvent &Call,
+ CheckerContext &C,
+ BinaryOperator::Opcode Op,
+ QualType ResultType) const {
+ // Calling a builtin with an incorrect argument count produces compiler error.
+ assert(Call.getNumArgs() == 3);
+
+ ProgramStateRef State = C.getState();
+ SValBuilder &SVB = C.getSValBuilder();
+ const Expr *CE = Call.getOriginExpr();
+
+ SVal Arg1 = Call.getArgSVal(0);
+ SVal Arg2 = Call.getArgSVal(1);
+
+ SVal RetValMax = SVB.evalBinOp(State, Op, Arg1, Arg2,
+ getSufficientTypeForOverflowOp(C, ResultType));
+ SVal RetVal = SVB.evalBinOp(State, Op, Arg1, Arg2, ResultType);
+
+ auto [Overflow, NotOverflow] = checkOverflow(C, RetValMax, ResultType);
+ if (NotOverflow) {
+ ProgramStateRef StateNoOverflow =
+ State->BindExpr(CE, C.getLocationContext(), SVB.makeTruthVal(false));
+
+ if (auto L = Call.getArgSVal(2).getAs<Loc>()) {
+ StateNoOverflow =
+ StateNoOverflow->bindLoc(*L, RetVal, C.getLocationContext());
+
+ // Propagate taint if any of the argumets were tainted
+ if (isTainted(State, Arg1) || isTainted(State, Arg2))
+ StateNoOverflow = addTaint(StateNoOverflow, *L);
+ }
+
+ C.addTransition(StateNoOverflow);
----------------
NagyDonat wrote:
> As far as I see, markInteresting is called on bug reported object. And there is no bug report object, since there is no bug here, we just do assumtions. I tried to find an example in other checkers, but I don't see it.
Yes, that's a good question -- the trick is that when you create a `NoteTag`, you can specify a callback that takes a `PathSensitiveBugReport` (by reference) and you should use `markInteresting()` within that callback. You could try to follow the example of `InvalidPtrChecker::createEnvInvalidationNote()` within `InvalidPtrChecker.cpp`, which implements something vaguely similar.
> (Like I see manipulations with markInteresting only during construction of bug report).
When a checker creates a bug report, it can use `markInteresting()` to mark the symbols that are _interesting_ (i.e. we used knowledge about them to conclude that the situation is buggy).
However, in addition to creating bug reports, checkers can also create `NoteTag`s when they do something interesting (e.g. make an assumption, model something important etc.) which is not a bug report. There are many trivial `NoteTag`s that just print the same plain text message on every bug report that passes through the execution step where the `NoteTag` was created, but in general the `NoteTag` may have an associated callback which inspects and modifies the `PathSensitiveBugReport` (in addition to emitting a "note" message).
The idea is that you should create a note tag with a callback which:
- checks whether the return value of the `__builtin_X_overflow()` is interesting and returns early if it isn't,
- marks the incoming arguments as interesting (if they're symbolic) to propagate interestingness;
- if both cases ("overflow" and "no overflow") were feasible, then returns a string message which shows the assumption that was selected.
(This is just a rough draft, there may be intricacies e.g. with describing taintedness.)
https://github.com/llvm/llvm-project/pull/102602
More information about the cfe-commits
mailing list