[Bug rtl-optimization/123322] New: [16 Regression] worse czero generation for GCC 16 vs GCC 15

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123322

            Bug ID: 123322
           Summary: [16 Regression] worse czero generation for GCC 16 vs
                    GCC 15
           Product: gcc
           Version: 16.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: rtl-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
  Target Milestone: ---
            Target: riscv

Take:
```
int f(int a, int b, int c)
{
  if (c) a |= b;
  return a;
}

```

Compile this with `-O2 -march=rv64gc_zicond -mabi=lp64d`. 

GCC 15 was able to produce:
```
        czero.eqz       a2,a1,a2
        or      a0,a0,a2
        ret
```
But GCC 16 produces:
```
        or      a1,a0,a1
        czero.eqz       a1,a1,a2
        czero.nez       a0,a0,a2
        add     a0,a0,a1
```

I think this was introduced by r16-3760-g9ff5cadac4579f (but I could be wrong).