[Bug tree-optimization/66423] New: a % (1 << b) no longer gets folded to a & (1 << b) for unsigned a and b

[Bug tree-optimization/66423] New: a % (1 << b) no longer gets folded to a & (1 << b) for unsigned a and b

[ppalka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66423

            Bug ID: 66423
           Summary: a % (1 << b) no longer gets folded to a & (1 << b) for
                    unsigned a and b
           Product: gcc
           Version: 6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ppalka at gcc dot gnu.org
  Target Milestone: ---

Test case:

unsigned int
my_mod (unsigned int a, unsigned int b)
{
    return a % (1 << b);
}

no longer gets folded to a & (1 << b).

I have not done a bisection but it may be that r223690 is the cause.

[Bug tree-optimization/66423] [6 Regression] a % (1 << b) no longer gets folded to a & (1 << b) for unsigned a

[glisse at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66423

Marc Glisse <glisse at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-06-05
   Target Milestone|---                         |6.0
            Summary|a % (1 << b) no longer gets |[6 Regression] a % (1 << b)
                   |folded to a & (1 << b) for  |no longer gets folded to a
                   |unsigned a and b            |& (1 << b) for unsigned a
     Ever confirmed|0                           |1

--- Comment #1 from Marc Glisse <glisse at gcc dot gnu.org> ---
a % (unsigned int) (1 << b)

The pattern in match.pd seems to be missing a 'convert' case.

[Bug tree-optimization/66423] [6 Regression] a % (1 << b) no longer gets folded to a & (1 << b) for unsigned a

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66423

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Mine.

[Bug tree-optimization/66423] [6 Regression] a % (1 << b) no longer gets folded to a & (1 << b) for unsigned a

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66423

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Tue Jun  9 12:31:43 2015
New Revision: 224279

URL: https://gcc.gnu.org/viewcvs?rev=224279&root=gcc&view=rev
Log:
2015-06-09  Richard Biener  <rguenther@suse.de>

        PR middle-end/66423
        * match.pd: Handle A % (unsigned)(1 << B).

        * gcc.dg/fold-modpow2.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/fold-modpow2.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/match.pd
    trunk/gcc/testsuite/ChangeLog