aaron.ballman added inline comments.
================
Comment at: clang/lib/AST/Interp/Interp.h:652-653
+ const SourceInfo &Loc = S.Current->getSource(OpPC);
+ S.FFDiag(Loc, diag::note_constexpr_pointer_comparison_unspecified)
+ << LS << RS;
+ return false;
----------------
tbaeder wrote:
> aaron.ballman wrote:
> > tbaeder wrote:
> > > aaron.ballman wrote:
> > > > Can we pass in the result of `getType()` instead of doing this string
> > > > conversion dance?
> > > Well the diagnostic doesn't print the result of the LHS/RHS:
> > > ```
> > > ./array.cpp:202:18: error: constexpr variable 'u13' must be initialized
> > > by a constant expression
> > > 202 | constexpr bool u13 = pf < pg; // ref-warning {{ordered
> > > comparison of function pointers}}
> > > | ^ ~~~~~~~
> > > ./array.cpp:202:27: note: comparison between '&f' and '&g' has
> > > unspecified value
> > > 202 | constexpr bool u13 = pf < pg; // ref-warning {{ordered
> > > comparison of function pointers}}
> > > | ^
> > > ```
> > >
> > > I'm not exactly a fan of how the code looks though. I might add a helper
> > > function for this later.
> > Ah of course, good point. And yeah, a helper function for this would
> > probably not be a bad idea.
> Do you like the `toDiagnosticString()` from https://reviews.llvm.org/D149172
> better?
I do, that's a nice helper. Feel free to either land those changes and use the
function here, or land these changes and do a later NFC commit to the helper
method.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D149154/new/
https://reviews.llvm.org/D149154
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
