[Bug tree-optimization/125748] New: `MAX` -> `a` should happen even if a is a signed type but nonnegative

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125748

            Bug ID: 125748
           Summary: `MAX<a, a&CST>` -> `a` should happen even if a is a
                    signed type but nonnegative
           Product: gcc
           Version: 16.0
            Status: UNCONFIRMED
          Keywords: easyhack, missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
  Target Milestone: ---

Take:
```
int
f (unsigned short a, unsigned short b) {
  int aa = a;
  aa++;
  int bb = aa & 0xFF;
  int t = aa > bb ? aa : bb;
  return t;
}

unsigned
g (unsigned short a, unsigned short b) {
  unsigned aa = a;
  aa++;
  unsigned bb = aa & 0xFF;
  unsigned t = aa > bb ? aa : bb;
  return t;
}
```
This 2 functions should produce the same code but currently does not since the
pattern:
```
(for minmax (min max)
 (simplify
  (minmax @0 (bit_and@1 @0 INTEGER_CST@2))
   (if (TYPE_UNSIGNED(type))
    (if (minmax == MIN_EXPR)
     @1
     @0))))
```

Only checks TYPE_UNSIGNED rather than @0 is nonnegative too.