[Bug middle-end/51428] New: Some code after SSA expand does nothing

[Bug middle-end/51428] New: Some code after SSA expand does nothing

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51428

             Bug #: 51428
           Summary: Some code after SSA expand does nothing
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: pinskia@gcc.gnu.org


While looking into PR 45416, I came across:
    /* Check for |= or &= of a bitfield of size one into another bitfield
       of size 1.  In this case, (unless we need the result of the
       assignment) we can do this more efficiently with a
       test followed by an assignment, if necessary.

>From what I am reading, this code does nothing any more as TREE_CODE (rhs) will
never be BIT_IOR_EXPR or BIT_AND_EXPR.

[Bug tree-optimization/51428] Some code after SSA expand does nothing

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51428

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |missed-optimization
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |pinskia at gcc dot gnu.org
          Component|middle-end                  |tree-optimization

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Mine for GCC 13 for bitfield lowering.
With the current code I have (which is applied to GCC 10 base compiler), we
get:
  _7 = MEM <unsigned char> [(struct f *)a_5(D)];
  _8 = _7 & 3;
  _3 = _8 != 0;
  _10 = BIT_INSERT_EXPR <_7, _3, 0 (1 bits)>;
  MEM <unsigned char> [(struct f *)a_5(D)] = _10;

And:
  _8 = _7 & 3;
  _6 = _8 == 3;
  _10 = BIT_INSERT_EXPR <_7, _6, 0 (1 bits)>;

both are not good really as we don't need to extract bit 0 really.