[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Richard Biener  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED
   Target Milestone|7.5 |8.0
  Known to fail||7.5.0

--- Comment #14 from Richard Biener  ---
Fixed in GCC8.

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Richard Biener  changed:

   What|Removed |Added

   Target Milestone|7.4 |7.5

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Richard Biener  changed:

   What|Removed |Added

   Target Milestone|7.3 |7.4

--- Comment #13 from Richard Biener  ---
GCC 7.3 is being released, adjusting target milestone.

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

--- Comment #12 from Richard Biener  ---
Author: rguenth
Date: Thu Sep 21 10:08:21 2017
New Revision: 253052

URL: https://gcc.gnu.org/viewcvs?rev=253052=gcc=rev
Log:
2017-09-21  Richard Biener  

PR tree-optimization/71351
* graphite-isl-ast-to-gimple.c (translate_isl_ast_to_gimple::
graphite_create_new_loop_guard): Remove, fold remaining parts
into caller ...
(translate_isl_ast_node_for): ... here and simplify.

* gfortran.dg/graphite/pr71351.f90: New testcase.
* gfortran.dg/graphite/interchange-3.f90: Adjust.

Added:
trunk/gcc/testsuite/gfortran.dg/graphite/pr71351.f90
Modified:
trunk/gcc/ChangeLog
trunk/gcc/graphite-isl-ast-to-gimple.c
trunk/gcc/testsuite/ChangeLog
trunk/gcc/testsuite/gfortran.dg/graphite/interchange-3.f90

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Richard Biener  changed:

   What|Removed |Added

  Known to work||8.0
Summary|[7/8 Regression] ICE:   |[7 Regression] ICE:
   |Segmentation fault  |Segmentation fault
   |(graphite)  |(graphite)

--- Comment #11 from Richard Biener  ---
Fixed on trunk.

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Richard Biener  changed:

   What|Removed |Added

 CC||spop at gcc dot gnu.org

--- Comment #8 from Richard Biener  ---
Also ICEs with -fgraphite-identity.

The issue seems to be that ISL creates a new loop guard but the loop has a
loop-close PHI node and we fail to generate/know the value to use on the
edge that skips the loop.  The condition we try to insert is _19 > 0
(that's trivially true by means of a dominating condition).

Not sure how this is supposed to work for reductions when the orginal loops
guard is not in the SESE region.  That guard looks like

  _19 = *nc_18(D);
  if (_19 <= 0)
...

so the BB is rejected because _19 = *nc_18(D) isn't a valid stmt.  So for this
case it might help if we'd split that block...

Anyway, it looks like we have to fail code generation here somehow... (no good
idea how).

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

--- Comment #7 from Richard Biener  ---
sese_insert_phis_for_liveouts looks like complete bollocks in this case.  We're
missing a definition that could be used for updating.

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Martin Liška  changed:

   What|Removed |Added

 CC||ktkachov at gcc dot gnu.org

--- Comment #6 from Martin Liška  ---
*** Bug 71142 has been marked as a duplicate of this bug. ***

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

--- Comment #5 from Martin Liška  ---
Just our of curiosity, why graphite does a transformation as original and AST
are same:

[scheduler] original ast:
{
  for (int c0 = 0; c0 < P_19; c0 += 1) {
S_4(c0);
for (int c1 = 0; c1 <= 2; c1 += 1)
  S_5(c0, c1);
S_7(c0);
  }
  S_9();
}

[scheduler] AST generated by isl:
{
  for (int c0 = 0; c0 < P_19; c0 += 1) {
S_4(c0);
for (int c1 = 0; c1 <= 2; c1 += 1)
  S_5(c0, c1);
S_7(c0);
  }
  S_9();
}

Doing a transformation should be easy in that case.

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

--- Comment #4 from Martin Liška  ---
Started with r236440.

[Bug tree-optimization/71351] [7 Regression] ICE: Segmentation fault (graphite)

[law at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71351

Jeffrey A. Law  changed:

   What|Removed |Added

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