[Bug tree-optimization/50328] New: loop interchange confuses vectorizer

[Bug tree-optimization/50328] New: loop interchange confuses vectorizer

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

             Bug #: 50328
           Summary: loop interchange confuses vectorizer
    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
                CC: irar@gcc.gnu.org


For

double dvec[256];

void test (void)
{
  long i, j;
  for (j = 0; j < 131072; ++j)
    for (i = 0; i < 256; ++i)
      dvec[i] *= 1.0000001;
}

the loops are interchanged with -Ofast -floop-interchange but the vectorizer
is confused by the extra IV lim inserts:

<bb 4>:
  # graphite_IV.6_21 = PHI <0(3), graphite_IV.6_22(5)>
  # dvec_I_lsm.7_26 = PHI <dvec_I_lsm.7_10(3), D.2732_25(5)>
  # ivtmp.9_19 = PHI <131072(3), ivtmp.9_29(5)>
  D.2732_25 = dvec_I_lsm.7_26 *
1.0000001000000000583867176828789524734020233154296875e+0;
  graphite_IV.6_22 = graphite_IV.6_21 + 1;
  ivtmp.9_29 = ivtmp.9_19 - 1;
  if (ivtmp.9_29 != 0)
    goto <bb 5>;
  else
    goto <bb 6>;

<bb 5>:
  goto <bb 4>;

this isn't detected as reduction for some reason.

[Bug tree-optimization/50328] loop interchange confuses vectorizer

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011-09-08
     Ever Confirmed|0                           |1

--- Comment #1 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-08 13:30:10 UTC ---
Doesn't seem to handle reduction with one operand being a constant.

[Bug tree-optimization/50328] loop interchange confuses vectorizer

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-08 13:36:23 UTC ---
With the following untested patch we apply outer loop vectorization.

Index: gcc/tree-vect-loop.c
===================================================================
--- gcc/tree-vect-loop.c        (revision 178687)
+++ gcc/tree-vect-loop.c        (working copy)
@@ -2149,7 +2149,7 @@ vect_is_simple_reduction_1 (loop_vec_inf
       op1 = gimple_assign_rhs1 (def_stmt);
       op2 = gimple_assign_rhs2 (def_stmt);

-      if (TREE_CODE (op1) != SSA_NAME || TREE_CODE (op2) != SSA_NAME)
+      if (TREE_CODE (op1) != SSA_NAME && TREE_CODE (op2) != SSA_NAME)
         {
           if (vect_print_dump_info (REPORT_DETAILS))
            report_vect_op (def_stmt, "reduction: uses not ssa_names: ");
@@ -2255,7 +2255,7 @@ vect_is_simple_reduction_1 (loop_vec_inf
     def2 = SSA_NAME_DEF_STMT (op2);

   if (code != COND_EXPR
-      && (!def1 || !def2 || gimple_nop_p (def1) || gimple_nop_p (def2)))
+      && ((!def1 && !def2) || (gimple_nop_p (def1) && gimple_nop_p (def2))))
     {
       if (vect_print_dump_info (REPORT_DETAILS))
        report_vect_op (def_stmt, "reduction: no defs for operands: ");
@@ -2268,6 +2268,7 @@ vect_is_simple_reduction_1 (loop_vec_inf

   if (def2 && def2 == phi
       && (code == COND_EXPR
+         || !def1
           || (def1 && flow_bb_inside_loop_p (loop, gimple_bb (def1))
               && (is_gimple_assign (def1)
                  || is_gimple_call (def1)
@@ -2285,6 +2286,7 @@ vect_is_simple_reduction_1 (loop_vec_inf

   if (def1 && def1 == phi
       && (code == COND_EXPR
+         || !def2
           || (def2 && flow_bb_inside_loop_p (loop, gimple_bb (def2))
              && (is_gimple_assign (def2)
                  || is_gimple_call (def2)

[Bug tree-optimization/50328] loop interchange confuses vectorizer

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

--- Comment #3 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-08 13:40:30 UTC ---
Triggered by report http://gcc.gnu.org/ml/gcc/2011-09/msg00052.html
OpenCC then unrolls the outer loop to get

.LBB16_double_array_mults_by_const:
 #<loop> Loop body line 62, nesting depth: 2, iterations: 16384
 #<loop> unrolled 4 times
        mulpd %xmm6,%xmm0               # [0]
        movaps %xmm0,%xmm1              # [4]
        mulpd %xmm6,%xmm1               # [6]
        mulpd %xmm6,%xmm1               # [10]
        addq $8,%rax                    # [14]
        mulpd %xmm6,%xmm1               # [14]
        cmpq $131071,%rax               # [15]
        setle %dil                      # [16]
        testb %dil,%dil                 # [17]
        movaps %xmm1,%xmm0              # [18]
        jne .LBB16_double_array_mults_by_const  # [18]

instead of what we get with the patch

.L2:
        subl    $1, %eax
        mulpd   %xmm1, %xmm0
        jne     .L2

we don't have outer loop unrolling either.

[Bug tree-optimization/50328] reduction with constant or invariant not vectorized

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

--- Comment #4 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-08 13:55:54 UTC ---
Created attachment 25228
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25228
preliminary patch

Patch that still ICEs its testcase.  Works for testcases that first require
loop interchange though.

[Bug tree-optimization/50328] reduction with constant or invariant not vectorized

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-08 14:06:58 UTC ---
Hmm, we fail to set the vectype for the PHI - I'm somewhat lost in the
reduction code.

[Bug tree-optimization/50328] reduction with constant or invariant not vectorized

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

--- Comment #6 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-09 09:05:39 UTC ---
Ah, we fail outer loop vectorization because of a bug (versioning for alias
required) and then drop into vectorizable_reduction with a
statement with a operand def that is a PHI with vect_unused_in_scope which of
course does not have its vector type set.  It's also not marked live
for some reason.  Doh, seems to be a genuine swap_tree_operands bug.

[Bug tree-optimization/50328] reduction with constant or invariant not vectorized

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

--- Comment #7 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-09 12:35:16 UTC ---
Author: rguenth
Date: Fri Sep  9 12:35:11 2011
New Revision: 178728

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=178728
Log:
2011-09-09  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/50328
    * tree-vect-loop.c (vect_is_simple_reduction_1): Allow one
    constant or default-def operand.

    * gcc.dg/vect/fast-math-vect-outer-7.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/vect/fast-math-vect-outer-7.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-vect-loop.c

[Bug tree-optimization/50328] reduction with constant or invariant not vectorized

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50328

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.7.0

--- Comment #8 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-09-09 12:36:29 UTC ---
Fixed for 4.7.