[PATCH] D87528: Enable '#pragma STDC FENV_ACCESS' in frontend cf. D69272 - Work in Progress

[Richard Smith - zygoloid via Phabricator via cfe-commits]

rsmith requested changes to this revision.
rsmith added inline comments.
This revision now requires changes to proceed.


================
Comment at: clang/lib/AST/ExprConstant.cpp:775-778
+    /// Strict floating point is enabled, this inhibits
+    /// floating ponit constant folding.
+    bool isStrictFP;
+
----------------
This is a property of the operation being constant-evaluated, not of the `EvalInfo`. The way you compute it below will only pick up the right value if the top-level expression in the evaluation happens to be a constrained floating-point evaluation.


================
Comment at: clang/lib/AST/ExprConstant.cpp:13197-13198
   assert(E->isRValue() && E->getType()->isRealFloatingType());
+  if (Info.isStrictFP)
+    return false;
   return FloatExprEvaluator(Info, Result).Visit(E);
----------------
I think we should be able to evaluate (for example) `constexpr float f = 1.0f;` even in a strict FP context. I think only floating point operations that depend on the rounding mode should be disabled, not all floating-point evaluation. Perhaps we should propagate the `FPOptions` into `handleFloatFloatBinOp` and perform the check there, along with any other places that care (I think we probably have some builtins that we can constant-evaluate that care about rounding modes.)

You also need to produce a diagnostic when you treat an expression as non-constant -- please read the comment at the top of the file for details.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D87528/new/

https://reviews.llvm.org/D87528