=?utf-8?q?Félix?= Cloutier <[email protected]>
Message-ID:
In-Reply-To: <llvm.org/llvm/llvm-project/pull/[email protected]>
================
@@ -2198,8 +2211,16 @@ llvm::Value *CodeGenFunction::EmitFromMemory(llvm::Value
*Value, QualType Ty) {
}
llvm::Type *ResTy = ConvertType(Ty);
- if (Ty->hasBooleanRepresentation() || Ty->isBitIntType() ||
- Ty->isExtVectorBoolType())
+ bool IsBitInt = Ty->isBitIntType();
+ bool HasBoolRep = Ty->hasBooleanRepresentation();
+ if (HasBoolRep && !IsBitInt &&
----------------
apple-fcloutier wrote:
If we treat `BitInt(1)` as a bool for the purposes of loading from memory, then:
* under `-fstrict-bool`, all BitInt values are truncated and insensitive to
their padding bits
* under `-fno-strict-bool=truncate`, all BitInt values are truncated and
insensitive to their padding bits
* under `-fno-strict-bool=nonzero`, all BitInt values are truncated and
insensitive to their padding bits _except_ for `BitInt(1)`, which considers the
value of its entire storage unit
There would be a strong case to make that it's a bug. Since BitInt values never
get !range metadata at this time, I think it's better to leave it alone.
If/when they do, there can be a `-f{,no-}strict-bitint` that controls how all
BitInt values, including BitInt(1), behave with respect to their padding bits.
https://github.com/llvm/llvm-project/pull/160790
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
