[Bug tree-optimization/103245] New: Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[Bug tree-optimization/103245] New: Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103245

            Bug ID: 103245
           Summary: Failure to detect abs pattern using multiplication
                    with slightly different casts for unsigned case
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
            Blocks: 103228
  Target Milestone: ---

While working on PR 103228, I got the testcase for PR 94785 failing.
It fails if I change the casts even:
unsigned int f4 (int v)
{
  unsigned int d_6;
  int b_5;
  int a_4;
  _Bool _1;
  unsigned int v1_2;
  unsigned int _7;
  int _9;

  _1 = v < 0;
  a_4 = (int) _1;
  b_5 = -a_4;
  _9 = b_5 | 1;
  d_6 = (unsigned int) _9;
  v1_2 = (unsigned int) v;
  _7 = v1_2 * d_6;
  return _7;
}
---- CUT ---


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103228
[Bug 103228] [9/10/11/12 Regression] missed optimization with |^ at the gimple
level

[Bug tree-optimization/103245] Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103245

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |pinskia at gcc dot gnu.org
   Last reconfirmed|                            |2021-11-15

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I will handle this tomorrow.

The current pattern matches:
(simplify
 (mult:c (bit_ior (nop_convert (negate (convert? (lt @0 integer_zerop))))
                 integer_onep) (nop_convert @0))

We need to match also:
(simplify
 (mult:c (nop_convert (bit_ior (negate (convert? (lt @0 integer_zerop)))
                                integer_onep) (nop_convert @0))

[Bug tree-optimization/103245] Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103245

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Created attachment 51801
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51801&action=edit
patch which I am testing

I noticed the patterns could be combined together.

[Bug tree-optimization/103245] Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103245

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |https://gcc.gnu.org/piperma
                   |                            |il/gcc-patches/2021-Novembe
                   |                            |r/584547.html
           Keywords|ra                          |patch

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Patch submitted:
https://gcc.gnu.org/pipermail/gcc-patches/2021-November/584547.html

[Bug tree-optimization/103245] Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103245

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The trunk branch has been updated by Andrew Pinski <pinskia@gcc.gnu.org>:

https://gcc.gnu.org/g:3200de91bc70bfd6d4b7b40e769bc82756b9b555

commit r12-5291-g3200de91bc70bfd6d4b7b40e769bc82756b9b555
Author: Andrew Pinski <apinski@marvell.com>
Date:   Mon Nov 15 22:55:09 2021 +0000

    tree-optimization: [PR103245] Improve detection of abs pattern using
multiplication

    So while working on PR 103228 (and a few others), I noticed the testcase
for PR 94785
    was failing. The problem is that the nop_convert moved from being inside
the IOR to be
    outside of it. I also noticed the patch for PR 103228 was not needed to
reproduce the
    issue either.
    This patch combines the two patterns together for the abs match when using
multiplication
    and adds a few places where nop_convert are optional.

    OK? Bootstrapped and tested on x86_64-linux-gnu with no regressions.

            PR tree-optimization/103245

    gcc/ChangeLog:

            * match.pd: Combine the abs pattern matching using multiplication.
            Adding optional nop_convert too.

    gcc/testsuite/ChangeLog:

            * gcc.dg/tree-ssa/pr103245-1.c: New test.

[Bug tree-optimization/103245] Failure to detect abs pattern using multiplication with slightly different casts for unsigned case

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103245

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED
   Target Milestone|---                         |12.0

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Fixed.