[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

--- Comment #14 from Richard Biener  ---
Author: rguenth
Date: Wed Feb  8 13:13:53 2017
New Revision: 245278

URL: https://gcc.gnu.org/viewcvs?rev=245278=gcc=rev
Log:
2017-02-08  Richard Biener  

Backport from mainline
2017-02-08  Richard Biener  

PR tree-optimization/71824
PR tree-optimization/79409
* gcc.dg/graphite/pr71824-3.c: New testcase.

2017-02-08  Richard Biener  

PR tree-optimization/71824
* graphite-scop-detection.c (scop_detection::build_scop_breadth):
Check all loops contained in the merged region.

* gcc.dg/graphite/pr71824-2.c: New testcase.

2017-02-01  Richard Biener  

PR tree-optimization/71824
* graphite-scop-detection.c (scop_detection::build_scop_breadth):
Verify the loops are valid in the merged SESE region.
(scop_detection::can_represent_loop_1): Check analyzing the
evolution of the number of iterations in the region succeeds.

* gcc.dg/graphite/pr71824.c: New testcase.

2017-01-31  Richard Biener  

PR tree-optimization/77318
* graphite-sese-to-poly.c (extract_affine): Fix assert.
(create_pw_aff_from_tree): Take loop parameter.
(add_condition_to_pbb): Pass loop of the condition to
create_pw_aff_from_tree.

Added:
branches/gcc-6-branch/gcc/testsuite/gcc.dg/graphite/pr71824-2.c
branches/gcc-6-branch/gcc/testsuite/gcc.dg/graphite/pr71824-3.c
branches/gcc-6-branch/gcc/testsuite/gcc.dg/graphite/pr71824.c
Modified:
branches/gcc-6-branch/gcc/ChangeLog
branches/gcc-6-branch/gcc/graphite-scop-detection.c
branches/gcc-6-branch/gcc/graphite-sese-to-poly.c
branches/gcc-6-branch/gcc/testsuite/ChangeLog

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Richard Biener  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED

--- Comment #13 from Richard Biener  ---
Fixed.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

--- Comment #12 from Richard Biener  ---
Author: rguenth
Date: Tue Jan 31 14:44:37 2017
New Revision: 245064

URL: https://gcc.gnu.org/viewcvs?rev=245064=gcc=rev
Log:
2017-01-31  Richard Biener  

PR tree-optimization/77318
* graphite-sese-to-poly.c (extract_affine): Fix assert.
(create_pw_aff_from_tree): Take loop parameter.
(add_condition_to_pbb): Pass loop of the condition to
create_pw_aff_from_tree.

Modified:
trunk/gcc/ChangeLog
trunk/gcc/graphite-sese-to-poly.c

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Richard Biener  changed:

   What|Removed |Added

 Status|NEW |ASSIGNED
   Assignee|unassigned at gcc dot gnu.org  |rguenth at gcc dot 
gnu.org

--- Comment #11 from Richard Biener  ---
Patch posted.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

--- Comment #10 from Richard Biener  ---
Ok, so the issue is we have a region that is inside of loop 1 but does not
fully cover it and in

/* Returns a linear expression for tree T evaluated in PBB.  */

static isl_pw_aff *
create_pw_aff_from_tree (poly_bb_p pbb, tree t)
{
  scop_p scop = PBB_SCOP (pbb);

  t = scalar_evolution_in_region (scop->scop_info->region, pbb_loop (pbb), t);

when analyzing _62 from a condition on a block inside the region (the blocks
loop father is loop 1) pbb_loop returns loop 3 (pbb->black_box->bb is its
loop header).

_62 is defined inside the region, but then

  if (TREE_CODE (t) != SSA_NAME
  || loop_in_sese_p (loop, region))
/* FIXME: we would need instantiate SCEV to work on a region, and be more
   flexible wrt. memory loads that may be invariant in the region.  */
return instantiate_scev (before, loop,
 analyze_scalar_evolution (loop, t));

here loop is loop 3 and thus we call analyze_scalar_evolution with bogus input
which just returns t, instantiate_scev then instantiates it to
(integer(kind=8)) (integer(kind=4)) {(unsigned int) pretmp_123 + 4294967295, +,
1}_1 but this has an evolution in a loop that is not in the region and boom.

Eventually that pbb is bogus from the start or rather that we are using pbb
when adding dominating conditions(?).

Index: gcc/graphite-sese-to-poly.c
===
--- gcc/graphite-sese-to-poly.c (revision 245052)
+++ gcc/graphite-sese-to-poly.c (working copy)
@@ -436,11 +436,11 @@ extract_affine (scop_p s, tree e, __isl_
 /* Returns a linear expression for tree T evaluated in PBB.  */

 static isl_pw_aff *
-create_pw_aff_from_tree (poly_bb_p pbb, tree t)
+create_pw_aff_from_tree (poly_bb_p pbb, loop_p loop, tree t)
 {
   scop_p scop = PBB_SCOP (pbb);

-  t = scalar_evolution_in_region (scop->scop_info->region, pbb_loop (pbb), t);
+  t = scalar_evolution_in_region (scop->scop_info->region, loop, t);

   gcc_assert (!chrec_contains_undetermined (t));
   gcc_assert (!automatically_generated_chrec_p (t));
@@ -455,8 +455,9 @@ create_pw_aff_from_tree (poly_bb_p pbb,
 static void
 add_condition_to_pbb (poly_bb_p pbb, gcond *stmt, enum tree_code code)
 {
-  isl_pw_aff *lhs = create_pw_aff_from_tree (pbb, gimple_cond_lhs (stmt));
-  isl_pw_aff *rhs = create_pw_aff_from_tree (pbb, gimple_cond_rhs (stmt));
+  loop_p loop = gimple_bb (stmt)->loop_father;
+  isl_pw_aff *lhs = create_pw_aff_from_tree (pbb, loop, gimple_cond_lhs
(stmt));
+  isl_pw_aff *rhs = create_pw_aff_from_tree (pbb, loop, gimple_cond_rhs
(stmt));

   isl_set *cond;
   switch (code)

fixes that but then we ICE in

#1  0x019bfc12 in extract_affine (s=0x2ab42a0, 
e=, space=0x2ae7790)
at /space/rguenther/src/svn/gcc-7-branch/gcc/graphite-sese-to-poly.c:409
408 case SSA_NAME:
409   gcc_assert (-1 != parameter_index_in_region_1 (e, s->scop_info)
410   || !invariant_in_sese_p_rec (e, s->scop_info->region,
NULL));

because the SSA name failed to register as parameter(?) - it is actually
defined inside!  But we run into

tree
scalar_evolution_in_region (const sese_l , loop_p loop, tree t)
{
...
  if (invariant_in_sese_p_rec (t, region, _vdefs))
return t;

which looks correct -- but it seems that extract_affine assumes we
"instantiate" an expression up to the regions params...

So maybe that assert is just another bogus one or scalar_evolution_in_region
is bogus (instantiating before the region entry returns a chrec with an
evolution in a loop not inside the region again...).

Well.  Doing

Index: gcc/graphite-sese-to-poly.c
===
--- gcc/graphite-sese-to-poly.c (revision 245052)
+++ gcc/graphite-sese-to-poly.c (working copy)
@@ -407,7 +407,7 @@ extract_affine (scop_p s, tree e, __isl_

 case SSA_NAME:
   gcc_assert (-1 != parameter_index_in_region_1 (e, s->scop_info)
- || !invariant_in_sese_p_rec (e, s->scop_info->region, NULL));
+ || defined_in_sese_p (e, s->scop_info->region));
   res = extract_affine_name (s, e, space);
   break;

fixes the testcase and still passes graphite.exp for all languages w/ {,-m32}.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[wschmidt at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

--- Comment #9 from Bill Schmidt  ---
Seen also on powerpc64-unknown-linux-gnu, but not on
powerpc64le-unknown-linux-gnu.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[danglin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

John David Anglin  changed:

   What|Removed |Added

 CC||danglin at gcc dot gnu.org

--- Comment #8 from John David Anglin  ---
Also seen on hppa64-hp-hpux11.11.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[law at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Jeffrey A. Law  changed:

   What|Removed |Added

   Priority|P3  |P4
 CC||law at redhat dot com

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Dominique d'Humieres  changed:

   What|Removed |Added

 CC||hjl.tools at gmail dot com

--- Comment #7 from Dominique d'Humieres  ---
*** Bug 71348 has been marked as a duplicate of this bug. ***

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[vries at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

--- Comment #6 from Tom de Vries  ---
In extract_affine_chrec we have:
...
  unsigned pos = sese_loop_depth (s->scop_info->region, get_chrec_loop (e)) -
1;
  isl_aff *loop = isl_aff_set_coefficient_si
(isl_aff_zero_on_domain (ls), isl_dim_in, pos, 1);
...

The immediate cause of the ICE is that sese_loop_depth returns 0, so pos ends
up with value UINT_MAX. This causes the out-of-bounds error during the call to
isl_aff_set_coefficient_si.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Dominique d'Humieres  changed:

   What|Removed |Added

 Target|aarch64-*-* |
 CC||vries at gcc dot gnu.org

--- Comment #5 from Dominique d'Humieres  ---
> It already failed at r236449 for both -m32 and -m64.

Likely r236440.

> It stopped to fail for -m32 between r236902 and r236930.

Reduced range from https://gcc.gnu.org/ml/gcc-regression/2016-05/msg00587.html:
between r236906 and r236909, likely 236907.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

--- Comment #4 from Dominique d'Humieres  ---
> It actually started between r236439 and r236491.

It already failed at r236449 for both -m32 and -m64. It stopped to fail for
-m32 between r236902 and r236930.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[sch...@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Andreas Schwab  changed:

   What|Removed |Added

 Blocks||70729

--- Comment #3 from Andreas Schwab  ---
Already noted in PR70729.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70729
[Bug 70729] Loop marked with omp simd pragma is not vectorized

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[sch...@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Andreas Schwab  changed:

   What|Removed |Added

 Blocks|77290   |

--- Comment #2 from Andreas Schwab  ---
It actually started between r236439 and r236491.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77290
[Bug 77290] [7 regression] test case gcc.dg/tree-ssa/pr71347.c fails starting
with r239565

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Dominique d'Humieres  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-08-22
 Ever confirmed|0   |1

--- Comment #1 from Dominique d'Humieres  ---
Confirmed. Note that the test has been failing since a long time up to revision
r239564 and passed only in the range r239565 to r239606.

[Bug tree-optimization/77318] [7 regression] FAIL: gfortran.dg/graphite/pr68279.f90 -O (internal compiler error)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77318

Richard Biener  changed:

   What|Removed |Added

   Target Milestone|--- |7.0