[Bug c++/116514] New: known pointer alignment is not used for constant propagation

[Bug c++/116514] New: known pointer alignment is not used for constant propagation

[mkretz at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116514

            Bug ID: 116514
           Summary: known pointer alignment is not used for constant
                    propagation
           Product: gcc
           Version: 15.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: mkretz at gcc dot gnu.org
  Target Milestone: ---

Test case (https://compiler-explorer.com/z/Gxd9KaG1T):

int
test(int* a, int* b)
{
  auto delta = (int*)__builtin_assume_aligned(b, 32)
               - (int*)__builtin_assume_aligned(a, 32);
  auto x = delta % 8;
  return (x == 0);
}

The expression 'delta % 8' should be optimized to a 0 constant. Consequently 'x
== 0' must satisfy __builtin_constant_p and be true. This could then be used to
completely avoid emitting code for a SIMD epilogue for given a range.

Clang manages to fold the above function to a constant 1.

[Bug c++/116514] known pointer alignment is not used for constant propagation

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116514

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
     Ever confirmed|0                           |1
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
   Last reconfirmed|                            |2024-08-28

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
CCP fails to handle

Visiting statement:
delta_7 = _3 /[ex] 4;
which is likely CONSTANT
Lattice value changed to VARYING.  Adding SSA edges to worklist.

so we get

Visiting statement:
x_8 = delta_7 % 8;
which is likely CONSTANT
Lattice value changed to VARYING.  Adding SSA edges to worklist.

despite

Visiting statement:
_3 = _1 - _2;
which is likely CONSTANT
Lattice value changed to CONSTANT 0x0 (0xffff...<1000s?!>...fffe0).

bit_value_binop supports TRUNC_DIV_EXPR but not EXACT_DIV_EXPR.

That's trivial to add.

[Bug c++/116514] known pointer alignment is not used for constant propagation

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116514

--- Comment #2 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Richard Biener <rguenth@gcc.gnu.org>:

https://gcc.gnu.org/g:e7d5b9aa021f6fc32810670b18ffabe543262775

commit r15-3548-ge7d5b9aa021f6fc32810670b18ffabe543262775
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Aug 28 14:06:48 2024 +0200

    tree-optimization/116514 - handle pointer difference in bit-CCP

    When evaluating the difference of two aligned pointers in CCP we
    fail to handle the EXACT_DIV_EXPR by the element size that occurs.
    The testcase then also exercises modulo to test alignment but
    modulo by a power-of-two isn't handled either.

            PR tree-optimization/116514
            * tree-ssa-ccp.cc (bit_value_binop): Handle EXACT_DIV_EXPR
            like TRUNC_DIV_EXPR.  Handle exact division of a signed value
            by a power-of-two like a shift.  Handle unsigned division by
            a power-of-two like a shift.
            Handle unsigned TRUNC_MOD_EXPR by power-of-two, handle signed
            TRUNC_MOD_EXPR by power-of-two if the result is zero.

            * gcc.dg/tree-ssa/ssa-ccp-44.c: New testcase.

[Bug c++/116514] known pointer alignment is not used for constant propagation

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116514

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED
   Target Milestone|---                         |15.0

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed for GCC 15.