[Bug c/52091] New: ICE: verify_gimple failed

[Bug c/52091] New: ICE: verify_gimple failed

[regehr at cs dot utah.edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

             Bug #: 52091
           Summary: ICE: verify_gimple failed
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: regehr@cs.utah.edu
                CC: chenyang@cs.utah.edu
              Host: x86_64-unknown-linux-gnu
            Target: x86_64-unknown-linux-gnu
             Build: x86_64-unknown-linux-gnu


[regehr@gamow tmp]$ current-gcc -v

Using built-in specs.
COLLECT_GCC=current-gcc
COLLECT_LTO_WRAPPER=/uusoc/exports/scratch/regehr/z/compiler-install/gcc-r183824-install/bin/../libexec/gcc/x86_64-unknown-linux-gnu/4.7.0/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../configure --with-libelf=/usr/local --enable-lto
--prefix=/home/regehr/z/compiler-install/gcc-r183824-install
--program-prefix=r183824- --enable-languages=c,c++
Thread model: posix
gcc version 4.7.0 20120202 (experimental) (GCC) 

[regehr@gamow tmp]$ current-gcc -Ofast -c small.c

small.c: In function 'fn52':
small.c:15:6: error: invalid operands in binary operation
vect_var_.33_31 = {pretmp.21_21, -1, -1, -1} & vect_cst_.34_29;

small.c:15:6: internal compiler error: verify_gimple failed
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

[regehr@gamow tmp]$ cat small.c

int fn39 (unsigned short p1)
{
  return p1;
}

int b,c,d,f;
unsigned h;

void fn58 (void)
{ 
  for (c = 0;;)
    return;
}

void fn52 (void)
{
  for (f = 4; f; f--)
    {
      h &= fn39 (d);
      for (b = 0; b <= 1; b++)
    {
      fn58 ();
      d &= 1;
    }
    }
}

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.7.0
            Summary|ICE: verify_gimple failed   |[4.7 Regression] ICE:
                   |                            |verify_gimple failed

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2012-02-02
               Host|x86_64-unknown-linux-gnu    |
     Ever Confirmed|0                           |1
              Build|x86_64-unknown-linux-gnu    |

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> 2012-02-02 04:53:25 UTC ---
Confirmed reduced testcase:
int b,c,d,f;
unsigned h;

void fn52 (void)
{
  for (f = 4; f; f--)
    {
      h &= (unsigned short)d;
      for (b = 0; b <= 1; b++)
      {
       c = 0;
       d &= 1;
      }
    }
}
---- CUT ----

Note also:
int b,c,d,f;
unsigned h;

void fn52 (void)
{
  for (f = 4; f; f--)
    {
      h &= d;
      for (b = 0; b <= 1; b++)
      {
       c = 0;
       d &= 1;
      }
    }
}
---- CUT ----
Gives a slightly different ICE:
t.c: In function ‘fn52’:
t.c:4:6: internal compiler error: in get_initial_def_for_reduction, at
tree-vect-loop.c:3391
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

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

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

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-02 08:29:41 UTC ---
Looking at it.

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-02 09:00:36 UTC ---
Started with http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=178728

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

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

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

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-02 12:06:27 UTC ---
Executable testcase:

/* PR tree-optimization/52091 */

int b, c, d, f;
unsigned h;
extern void abort (void);

int
main ()
{
  d = -1;
  h = 65;
  asm volatile ("" : : : "memory");
  for (f = 0; f < 4; f++)
    {
      h &= (unsigned short) d;
      for (b = 0; b <= 1; b++)
    {
      c = 0;
      d &= 1;
    }
    }
  asm volatile ("" : : : "memory");
  if (b != 2 || c != 0 || d != 1 || f != 4 || h != 1)
    abort ();
  return 0;
}

Following patch fixes the ICE:
--- gcc/tree-vect-stmts.c.jj    2012-02-01 10:33:58.000000000 +0100
+++ gcc/tree-vect-stmts.c    2012-02-02 11:50:42.623948066 +0100
@@ -1294,7 +1294,11 @@ vect_get_vec_def_for_operand (tree op, g

         /* Get the def before the loop  */
         op = PHI_ARG_DEF_FROM_EDGE (def_stmt, loop_preheader_edge (loop));
-        return get_initial_def_for_reduction (stmt, op, scalar_def);
+        vec_oprnd = get_initial_def_for_reduction (stmt, op, scalar_def);
+        if (!is_gimple_val (vec_oprnd))
+      vec_oprnd = vect_init_vector (stmt, vec_oprnd,
+                    TREE_TYPE (vec_oprnd), NULL);
+    return vec_oprnd;
      }

     /* Case 5: operand is defined by loop-header phi - induction.  */
but just turns ice-on-valid-code into wrong-code, h is 65 when vectorizing.

Ira, could you please look at this?  Thanks.

OT, it is strange that we are creating a reduction for a loop which loops
exactly as many times as there are units in the vector, that doesn't seem to be
profitable.

My other attempt to fix the ICE was:
--- gcc/tree-vect-loop.c.jj    2011-12-16 08:37:45.000000000 +0100
+++ gcc/tree-vect-loop.c    2012-02-02 11:57:25.354607843 +0100
@@ -3370,7 +3370,10 @@ get_initial_def_for_reduction (gimple st
         if (TREE_CONSTANT (init_val))
           init_def = build_vector (vectype, t);
         else
-          init_def = build_constructor_from_list (vectype, t);
+      {
+        init_def = build_constructor_from_list (vectype, t);
+        init_def = vect_init_vector (stmt, init_def, vectype, NULL);
+      }

         break;
but that ICEs on the testcase, because the other caller of
get_initial_def_for_reduction is calling vect_init_vector immediately after it,
with a different stmt than was passed to get_initial_def_for_reduction, and
this resulted in definition not dominating the use.  Wonder if that isn't
another bug.

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

Ira Rosen <irar at il dot ibm.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |irar at il dot ibm.com

--- Comment #5 from Ira Rosen <irar at il dot ibm.com> 2012-02-02 13:22:53 UTC ---
(In reply to comment #4)

I think that the problem here is that 
D.2030_19 = prephitmp.17_21 & 65535
is falsely initialized as reduction, while it isn't a reduction. The mistake
occurs because its def_stmt, 
prephitmp.17_21 = PHI <d.3_44(7), pretmp.16_28(2)> ,
is a phi node marked as double reduction:

<bb 3>:
  # f.6_36 = PHI <f.7_13(7), 0(2)>
  # prephitmp.17_21 = PHI <d.3_44(7), pretmp.16_28(2)> - double reduction phi
   ...
  D.2030_19 = prephitmp.17_21 & 65535; - not a reduction stmt
   ...
<bb 4>:
  # b.4_41 = PHI <b.5_11(5), 0(3)>
   ...
  d.3_9 = d_lsm.23_31 & 1;
   ...

<bb 6>:
  # d.3_44 = PHI <d.3_9(4)>       - double reduction stmt

I think we should fail to vectorize D.2030_19 = prephitmp.17_21 & 65535, or any
other non-phi/not vect_double_reduction_def stmt with a double reduction phi as
a def_stmt.

We can either check this in every vectorizable_* for every operand, like this:
Index: tree-vect-stmts.c
===================================================================
--- tree-vect-stmts.c   (revision 183125)
+++ tree-vect-stmts.c   (working copy)
@@ -3326,7 +3326,8 @@ vectorizable_operation (gimple stmt, gimple_stmt_i

   op0 = gimple_assign_rhs1 (stmt);
   if (!vect_is_simple_use_1 (op0, loop_vinfo, bb_vinfo,
-                            &def_stmt, &def, &dt[0], &vectype))
+                            &def_stmt, &def, &dt[0], &vectype)
+      || dt[0] == vect_double_reduction_def)
     {
       if (vect_print_dump_info (REPORT_DETAILS))
         fprintf (vect_dump, "use not simple.");


or pass stmt or stmt_info to vect_is_simple_use and check it there.


> OT, it is strange that we are creating a reduction for a loop which loops
> exactly as many times as there are units in the vector, that doesn't seem to be
> profitable.
> 

Right, but doesn't cost model catch this?

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-02 13:41:01 UTC ---
(In reply to comment #5)
> I think we should fail to vectorize D.2030_19 = prephitmp.17_21 & 65535, or any
> other non-phi/not vect_double_reduction_def stmt with a double reduction phi as
> a def_stmt.
> 
> We can either check this in every vectorizable_* for every operand, like this:
> Index: tree-vect-stmts.c
> ===================================================================
> --- tree-vect-stmts.c   (revision 183125)
> +++ tree-vect-stmts.c   (working copy)
> @@ -3326,7 +3326,8 @@ vectorizable_operation (gimple stmt, gimple_stmt_i
> 
>    op0 = gimple_assign_rhs1 (stmt);
>    if (!vect_is_simple_use_1 (op0, loop_vinfo, bb_vinfo,
> -                            &def_stmt, &def, &dt[0], &vectype))
> +                            &def_stmt, &def, &dt[0], &vectype)
> +      || dt[0] == vect_double_reduction_def)
>      {
>        if (vect_print_dump_info (REPORT_DETAILS))
>          fprintf (vect_dump, "use not simple.");
> 
> 
> or pass stmt or stmt_info to vect_is_simple_use and check it there.

Are you going to write a patch for this?  Not sure how exactly would you like
it to look up.

> > OT, it is strange that we are creating a reduction for a loop which loops
> > exactly as many times as there are units in the vector, that doesn't seem to be
> > profitable.
> > 
> 
> Right, but doesn't cost model catch this?

For simple testcases it does apparently.

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

--- Comment #7 from Ira Rosen <irar at il dot ibm.com> 2012-02-02 15:16:18 UTC ---
(In reply to comment #6)
> (In reply to comment #5)
> > I think we should fail to vectorize D.2030_19 = prephitmp.17_21 & 65535, or any
> > other non-phi/not vect_double_reduction_def stmt with a double reduction phi as
> > a def_stmt.
> > 
> > We can either check this in every vectorizable_* for every operand, like this:
> > Index: tree-vect-stmts.c
> > ===================================================================
> > --- tree-vect-stmts.c   (revision 183125)
> > +++ tree-vect-stmts.c   (working copy)
> > @@ -3326,7 +3326,8 @@ vectorizable_operation (gimple stmt, gimple_stmt_i
> > 
> >    op0 = gimple_assign_rhs1 (stmt);
> >    if (!vect_is_simple_use_1 (op0, loop_vinfo, bb_vinfo,
> > -                            &def_stmt, &def, &dt[0], &vectype))
> > +                            &def_stmt, &def, &dt[0], &vectype)
> > +      || dt[0] == vect_double_reduction_def)
> >      {
> >        if (vect_print_dump_info (REPORT_DETAILS))
> >          fprintf (vect_dump, "use not simple.");
> > 
> > 
> > or pass stmt or stmt_info to vect_is_simple_use and check it there.
> 
> Are you going to write a patch for this?  Not sure how exactly would you like
> it to look up.
> 

We need to check that if def_stmt is vect_double_reduction_def, the stmt itself
needs to be vect_double_reduction_def. We know def_type (dt in the above patch)
of def_stmt from vect_is_simple_use. We call it from all the vectorizable_*
functions. We don't pass the stmt itself to vect_is_simple_use, therefore, we
should either do that and perform the check in vect_is_simple_use, or check
this in all the vectorizable_* (except perhaps vectorizable_reduction). I
prefer to pass stmt or stmt_info to vect_is_simple_use.

I can do this on Sunday. You are welcome to do this yourself. Whatever you
prefer.

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[irar at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

--- Comment #8 from irar at gcc dot gnu.org 2012-02-05 05:58:22 UTC ---
Author: irar
Date: Sun Feb  5 05:58:18 2012
New Revision: 183902

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=183902
Log:

        PR tree-optimization/52091
        * tree-vectorizer.h (vect_is_simple_use): Add an argument.
        (vect_is_simple_use_1): Likewise.
        * tree-vect-loop.c (vectorizable_reduction): Update calls
        to vect_is_simple_use_1 and vect_is_simple_use.
        (vectorizable_live_operation): Likewise.
        * tree-vect-patterns.c (widened_name_p,
        vect_recog_vector_vector_shift_pattern, check_bool_pattern):
        Likewise.
        * tree-vect-stmts.c (process_use, vect_get_vec_def_for_operand,
        vectorizable_call, vectorizable_conversion,
        vectorizable_assignment, vectorizable_shift,
        vectorizable_operation, vectorizable_store, vectorizable_load):
        Likewise.
        (vect_is_simple_cond): Add an argument, pass it to
        vect_is_simple_use_1.
        (vectorizable_condition): Update calls to vect_is_simple_cond,
        vect_is_simple_use.
        (vect_is_simple_use): Add an argument, the statement in which
        OPERAND is used.  Check that if OPERAND's def stmt is a double
        reduction phi node, the use is a phi node too.
        (vect_is_simple_use_1): Add an argument, pass it to
        vect_is_simple_use.
        * tree-vect-slp.c (vect_get_and_check_slp_defs): Update a call
        to vect_is_simple_use.


Added:
    trunk/gcc/testsuite/gcc.dg/vect/pr52091.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-vect-loop.c
    trunk/gcc/tree-vect-patterns.c
    trunk/gcc/tree-vect-slp.c
    trunk/gcc/tree-vect-stmts.c
    trunk/gcc/tree-vectorizer.h

[Bug tree-optimization/52091] [4.7 Regression] ICE: verify_gimple failed

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52091

Ira Rosen <irar at il dot ibm.com> changed:

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

--- Comment #9 from Ira Rosen <irar at il dot ibm.com> 2012-02-06 08:42:00 UTC ---
Fixed.