aaron.ballman added inline comments.
================
Comment at: clang/lib/AST/Interp/Boolean.h:113
+ static Boolean bitcastFromMemory(const std::byte *Buff) {
+ bool Val = static_cast<bool>(*Buff);
+ return Boolean(Val);
----------------
tbaeder wrote:
> MitalAshok wrote:
> > Does this handle padding bits correctly? E.g., `__builtin_bit_cast(bool,
> > (unsigned char) 0b10u)` should be false
> No, it doesn't. But clang has a completely different opinion about that test.
I think Clang diagnoses that as an invalid constant expression because of
https://eel.is/c++draft/utilities#bit.cast-2.sentence-5 which we're allowed to
do because of http://eel.is/c++draft/expr.const#5.31
================
Comment at: clang/lib/AST/Interp/Interp.h:1546
+ const Pointer &FromPtr = S.Stk.pop<Pointer>();
+ Pointer &ToPtr = S.Stk.peek<Pointer>();
+
----------------
Should we be checking that we don't have a member pointer at this point?
================
Comment at: clang/lib/AST/Interp/InterpBitcast.cpp:19
+
+// TODO: Try to e-duplicate the primitive and composite versions.
+
----------------
Alternatively: `e-duplicate2000`
================
Comment at: clang/lib/AST/Interp/InterpBitcast.cpp:71
+/// All offsets are in bytes.
+struct ByteTracker {
+ std::vector<bool> Initialized;
----------------
Don't we need to track this at the *bit* level instead of the *byte* level?
e.g., padding bits in structures, anonymous bit-fields, `bool`, `_BitInt`, etc?
================
Comment at: clang/lib/AST/Interp/InterpBitcast.cpp:72
+struct ByteTracker {
+ std::vector<bool> Initialized;
+ std::vector<std::byte> Data;
----------------
`llvm::BitVector` instead of `std::vector<bool>`?
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D154951/new/
https://reviews.llvm.org/D154951
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
