[Bug tree-optimization/102872] New: If statement is always false but not figured out at gimple level

[Bug tree-optimization/102872] New: If statement is always false but not figured out at gimple level

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

            Bug ID: 102872
           Summary: If statement is always false but not figured out at
                    gimple level
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Keywords: missed-optimization, TREE
          Severity: enhancement
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
  Target Milestone: ---

Take:
void foo(void);

static int a, b;
int main() {
  for (; a; ++a) {
    unsigned short d = a;
    if (!(b | d) && d)
      foo();
  }
}
---- CUT ---
This is the reduced testcase of PR 102703 but without the store to a static
variable which then was removed.

In VRP1 we have:
  d_10 = (short unsigned int) a.3_5;
  _9 = a.3_5 & 65535;
  if (_9 == 0)
    goto <bb 4>; [50.00%]
  else
    goto <bb 6>; [50.00%]

  <bb 4> [local count: 477815112]:
  if (d_10 != 0)
    goto <bb 5>; [33.00%]
  else
    goto <bb 6>; [67.00%]

  <bb 5> [local count: 157678986]:
  foo ();

_9 and d_10 have the same value though different types.

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Note this does not get optimized at the rtl level on all targets either, e.g.
aarch64 we get:
        tst     w0, 65535
        and     w0, w0, 65535
        ccmp    w0, 0, 4, eq
        beq     .L3
        bl      foo

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-10-21
                 CC|                            |amacleod at redhat dot com,
                   |                            |rguenth at gcc dot gnu.org

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed.  I wonder if the relation oracle could be extended to handle this
though there's the usual caveat that _9 and d_10 are not dependent on each
other but only are indirectly related through a.3_5.

  d_10 = (short unsigned int) a.3_5;
  _9 = a.3_5 & 65535;
  if (_9 == 0)

Here VN could be teached to look for (short unsigned int) a.3_5 when
seeing a.3_5 & 65535 and CSE it as (short signed int) d_10 (but then
I think we might fold that back).

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I wonder if we should just do this:
(for neeq (ne eq)
(simplify
 (neeq (convert:s @0) integer_zerop@1)
   (with
    {
      tree inside_type = TREE_TYPE (@0);
      tree outer_type = TREE_TYPE (@1);
    }
    (if (TYPE_PRECISION (inside_type) > TYPE_PRECISION (outer_type))
     (neeq (bit_and @0
            { wide_int_to_tree (type0,
                                wi::mask (TYPE_PRECISION (inside_type),
                                          false,
                                          TYPE_PRECISION (outer_type))); })
             { build_zero_cst (inside_type); })))))
)

Or does the bit_and to a convert make better sense? I don't care either way
really but it would be a good idea to chose one or the other really.

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
ifcombine makes

optimizing two comparisons to _13
Merging blocks 3 and 4

to

  <bb 3> [local count: 955630225]:
  d_10 = (short unsigned int) a.3_5;
  _9 = a.3_5 & 65535;
  _2 = d_10 != 0;
  _1 = _9 == 0;
  _13 = _1 & _2;
  if (_13 != 0)

so I wonder if we can simplify that somehow.  It looks like this could be
simplified to (a & 0xffff0000) != 0?

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

--- Comment #5 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Andrew Macleod <amacleod@gcc.gnu.org>:

https://gcc.gnu.org/g:6cc3394507a2303a18891d34222c53f679256c37

commit r13-3281-g6cc3394507a2303a18891d34222c53f679256c37
Author: Andrew MacLeod <amacleod@redhat.com>
Date:   Wed Oct 5 10:42:07 2022 -0400

    propagate partial equivs in the cache.

    Adjust on-entry cache propagation to look for and propagate both full
    and partial equivalences.

            gcc/
            PR tree-optimization/102540
            PR tree-optimization/102872
            * gimple-range-cache.cc (ranger_cache::fill_block_cache):
            Handle partial equivs.
            (ranger_cache::range_from_dom): Cleanup dump output.

            gcc/testsuite/
            * gcc.dg/pr102540.c: New.
            * gcc.dg/pr102872.c: New.

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[amacleod at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

Andrew Macleod <amacleod at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|NEW                         |RESOLVED

--- Comment #6 from Andrew Macleod <amacleod at redhat dot com> ---
fixed.

[Bug tree-optimization/102872] If statement is always false but not figured out at gimple level

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102872

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |13.0