[Bug tree-optimization/51692] New: [4.7 Regression] ICE on several valgrind tests

[Bug tree-optimization/51692] New: [4.7 Regression] ICE on several valgrind tests

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

             Bug #: 51692
           Summary: [4.7 Regression] ICE on several valgrind tests
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: jakub@gcc.gnu.org
            Target: x86_64-linux


int
main ()
{
  volatile double d = 0.0;
  double *p = __builtin_calloc (1, sizeof (double));
  d += 1.0;
  *p += 2.0;
  __builtin_free (p);
  return 0;
}

ICEs at -O2, the free argument becomes a freed SSA_NAME for some reason.
Started with http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=182009

[Bug tree-optimization/51692] [4.7 Regression] ICE on several valgrind tests

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.7.0

[Bug tree-optimization/51692] [4.7 Regression] ICE on several valgrind tests

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011-12-30
     Ever Confirmed|0                           |1

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-12-30 22:28:54 UTC ---
Confirmed.
(gdb) bt
#0  is_gimple_reg_type (type=0x0) at
/home/pinskia/src/local/gcc/gcc/gimple.c:2916
#1  0x00000000008c12b0 in verify_gimple_call (stmt=0x7ffff6d3c5d8)
    at /home/pinskia/src/local/gcc/gcc/tree-cfg.c:3229
#2  0x00000000008c89ef in verify_gimple_in_cfg (fn=0x7ffff6d3f140)
    at /home/pinskia/src/local/gcc/gcc/tree-cfg.c:4520
#3  0x0000000000a11ef6 in verify_ssa (check_modified_stmt=true) at
/home/pinskia/src/local/gcc/gcc/tree-ssa.c:928
#4  0x00000000008034ec in execute_function_todo (data=<optimized out>)
    at /home/pinskia/src/local/gcc/gcc/passes.c:1725
#5  0x0000000000803e6e in execute_todo (flags=34862) at
/home/pinskia/src/local/gcc/gcc/passes.c:1756
#6  0x000000000080705a in execute_one_pass (pass=pass@entry=0x13a7f80)
    at /home/pinskia/src/local/gcc/gcc/passes.c:2102

[Bug tree-optimization/51692] [4.7 Regression] ICE on several valgrind tests

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-12-30 22:57:15 UTC ---
I think the problem is that for BUILT_IN_FREE we don't mark_operand_necessary,
but then the BUILT_IN_CALLOC call is marked as necessary anyway through
mark_aliased_reaching_defs_necessary_1 calling mark_operand_necessary (vdef).
The *p += 2.0; stmts are removed and for the calloc eliminate_unnecessary_stmts
performs:
                  /* When LHS of var = call (); is dead, simplify it into
                     call (); saving one operand.  */
and thus drops the lhs SSA_NAME, eventhough it is still used by the free call.
Richard?

[Bug tree-optimization/51692] [4.7 Regression] ICE on several valgrind tests

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

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

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

--- Comment #3 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-01-02 10:26:37 UTC ---
Mine.

[Bug tree-optimization/51692] [4.7 Regression] ICE on several valgrind tests

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

--- Comment #4 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-01-03 11:54:58 UTC ---
Author: rguenth
Date: Tue Jan  3 11:54:53 2012
New Revision: 182838

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=182838
Log:
2012-01-03  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/51692
    * tree-ssa-dce.c (eliminate_unnecessary_stmts): Do not remove
    the LHS of allocation stmts.

    * gcc.dg/torture/pr51692.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/torture/pr51692.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-ssa-dce.c

[Bug tree-optimization/51692] [4.7 Regression] ICE on several valgrind tests

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51692

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

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

--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-01-03 11:55:21 UTC ---
Fixed.