https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114084
--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Here is another testcase, the only use of _BitInt is in the cast, everything
else is a bitfield:
```
struct a
{
unsigned t:(sizeof(int)*8-1);
};
typedef unsigned _BitInt(sizeof(int)*8-1) ub31;
struct a ub63_0;
struct a ub63_1;
void
foo (void)
{
ub63_1.t = (ub31)
((ub63_0.t | (-1u>>1) ) >> 1 | (ub63_0.t | 1u) << 4);
}
```
I still can't figure out why the cast is needed though.
