https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121736
Bug ID: 121736
Summary: Offsetof with non-constant array index should be
diagnosed as an extension
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: luigighiron at gmail dot com
Target Milestone: ---
The following code is accepted by GCC even with -pedantic-errors:
#include<stddef.h>
int main(){
int x=2;
struct y{
int z[4];
};
offsetof(struct y,z[x]);
}
This code shouldn't be accepted with -pedantic-errors because it violates the
following text:
> The macros are
>
> ...
>
> offsetof(type, member-designator)
>
> which expands to an integer constant expression that has type size_t, the
> value of which is the offset in bytes, to the subobject (designated by
> member-designator), from the beginning of any object of type type. The type
> and member designator shall be such that given
>
> static type t;
>
> then the expression &(t.member-designator) evaluates to an address constant.
> If the specified type name contains a comma not between matching parentheses
> or if the specified member is a bit-field, the behavior is undefined.
Section 7.21 "Common definitions <stddef.h>" Paragraph 4 ISO/IEC 9899:2024
Strictly, the behavior is undefined so not diagnosing this is correct. However,
diagnosing it with -pedantic seems simple to do so I think it should be done.
As an aside, the GCC C++ compiler doesn't support this extension. Clang and
MSVC support this extension in both C and C++ (though they also don't diagnose
it).
