[PATCH] D87822: [FPEnv] Evaluate constant expressions under non-default rounding modes

[John McCall via Phabricator via cfe-commits]

rjmccall accepted this revision.
rjmccall added inline comments.
This revision is now accepted and ready to land.


================
Comment at: clang/lib/AST/ExprConstant.cpp:2427
+    FPOptions FPFeatures = Cast->getFPFeaturesInEffect(Info.Ctx.getLangOpts());
+    RM = FPFeatures.getRoundingMode();
+  }
----------------
sepavloff wrote:
> rjmccall wrote:
> > sepavloff wrote:
> > > rjmccall wrote:
> > > > I think the options really need to be passed in or else correctness is somewhat doomed here.
> > > > 
> > > > For example, the call to CompoundAssignSubobjectHandler needs to propagate this down from the operator expression.
> > > It is guaranteed by the way AST is built, no?
> > > 
> > > As FP options may be changed only by pragmas and the pragmas can be specified only at file or block level, all sub-expression are evaluated at the same options.
> > Yes, but you can't actually reliably recover those settings from E unless you're sure E is one of a few kinds of expression.  The concern is that E might end up being some closely-related expression that isn't actually the expression that carries the current settings, and then we'll fall back on using the global defaults.  It's much more correct-by-construction to pass the settings down from the caller based on the caller's static knowledge of which expression is under consideration, and I think you'll see that that's pretty straightforward in practice.
> This is a peculiarity of `CompoundAssignOperator`, which itself makes conversion, without `CastExpr`. I added `assert` to ensure that other nodes cannot appear here. Also some tests with conversion in `CompoundAssignOperator` were added.
Alright, well, I don't understand the reluctance to pass this down, but I won't insist.  LGTM.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D87822/new/

https://reviews.llvm.org/D87822