[Bug tree-optimization/52019] New: [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[Bug tree-optimization/52019] New: [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

             Bug #: 52019
           Summary: [4.7 Regression] tree-ssa/ipa-split-5.c fails with
                    -fno-tree-sra because of CLOBBERS
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: rguenth@gcc.gnu.org


IPA split is confused by the stale clobber in BB 5:

<bb 2>:
  D.1717_2 = a_1(D) != 0;
  D.1718_3 = (long int) D.1717_2;
  D.1719_4 = __builtin_expect (D.1718_3, 1);
  if (D.1719_4 != 0)
    goto <bb 3>;
  else
    goto <bb 4>;

<bb 3>:
  D.1722.a = 0;
  D.1722.b = 0;
  goto <bb 5>;

<bb 4>:
  retval = make_me_big (a_1(D));
  retval = make_me_big (a_1(D));
  retval = make_me_big (a_1(D));
  retval = make_me_big (a_1(D));
  retval = make_me_big (a_1(D));
  retval = make_me_big (a_1(D));
  D.1722 = retval;
  retval ={v} {CLOBBER};

<bb 5>:
  retval ={v} {CLOBBER};
  return D.1722;


As you can see 'retval' is completely unused on the path to BB5.  We
should arrange for them to be removed in such cases, maybe by DCE.

This blocks SRA from not "scalarizing" clobbers, which it does like:

<bb 5>:
  retval$a_21 = 0;
  retval$b_22 = 0;
  return D.1722;

which is sort-of bogus anyway.

I'll fixup SRA, but for SRA to do this trick there needs to be stmts
that touch retval in any meaningful way, so a clobber that is stale
before SRA will stay so.

[Bug tree-optimization/52019] [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |matz at gcc dot gnu.org
   Target Milestone|---                         |4.7.0

[Bug tree-optimization/52019] [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-22 10:11:28 UTC ---
Shouldn't IPA split just ignore clobbers when deciding whether to split?
Then obviously it needs to handle them somehow, but IMHO the decision shouldn't
take them into account.

[Bug tree-optimization/52019] [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[rguenther at suse dot de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

--- Comment #2 from rguenther at suse dot de <rguenther at suse dot de> 2012-02-22 10:16:01 UTC ---
On Wed, 22 Feb 2012, jakub at gcc dot gnu.org wrote:

> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019
> 
> Jakub Jelinek <jakub at gcc dot gnu.org> changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>                  CC|                            |jakub at gcc dot gnu.org
> 
> --- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-22 10:11:28 UTC ---
> Shouldn't IPA split just ignore clobbers when deciding whether to split?
> Then obviously it needs to handle them somehow, but IMHO the decision shouldn't
> take them into account.

Yes.  The question is how exactly ... the idea was that dropping them
(for 'unused' vars) would be appropriate.

Richard.

[Bug tree-optimization/52019] [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2012-02-23
         AssignedTo|unassigned at gcc dot       |jakub at gcc dot gnu.org
                   |gnu.org                     |
     Ever Confirmed|0                           |1

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-23 11:36:43 UTC ---
Created attachment 26726
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26726
gcc47-pr52019.patch

Untested fix.

[Bug tree-optimization/52019] [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-23 14:43:48 UTC ---
Author: jakub
Date: Thu Feb 23 14:43:43 2012
New Revision: 184508

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=184508
Log:
    PR tree-optimization/52019
    * ipa-split.c (find_return_bb, find_retval, visit_bb): Ignore
    CLOBBER stmts.

    * gcc.dg/tree-ssa/ipa-split-6.c: New test.

Added:
    trunk/gcc/testsuite/gcc.dg/tree-ssa/ipa-split-6.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/ipa-split.c
    trunk/gcc/testsuite/ChangeLog

[Bug tree-optimization/52019] [4.7 Regression] tree-ssa/ipa-split-5.c fails with -fno-tree-sra because of CLOBBERS

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52019

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

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

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-23 15:19:58 UTC ---
Fixed.