[Bug tree-optimization/46723] New: internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[Bug tree-optimization/46723] New: internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[kretz at kde dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

           Summary: internal compiler error: in
                    get_initial_def_for_induction, at
                    tree-vect-loop.c:2431
           Product: gcc
           Version: 4.5.1
            Status: UNCONFIRMED
          Severity: critical
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: kretz@kde.org


Compile the attached (preprocessed) source file with
/opt/gcc-4.5.1/bin/g++ -O3 -o tmp.o -c gcc_4.5.1_ice.cpp

and you'll get
/home/mkretz/src/testing/Vc-Nightly/tests/memory.cpp: In static member function
‘static void TestVectors<V, Size>::test() [with V = Vc::Simple::Vector<short
int>, unsigned int Size = 128u]’:
/home/mkretz/src/testing/Vc-Nightly/tests/memory.cpp:89:46: internal compiler
error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[Bug tree-optimization/46723] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[kretz at kde dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #1 from Matthias Kretz <kretz at kde dot org> 2010-11-30 13:50:55 UTC ---
Created attachment 22578
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22578
preprocessed source which makes G++ 4.5.1 ICE

[Bug tree-optimization/46723] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2010.11.30 14:27:37
                 CC|                            |irar at gcc dot gnu.org,
                   |                            |rguenth at gcc dot gnu.org
      Known to work|                            |4.5.0
   Target Milestone|---                         |4.5.2
     Ever Confirmed|0                           |1
      Known to fail|                            |4.5.1

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-11-30 14:27:37 UTC ---
Confirmed.  This is a regression on the branch, trunk works, so maybe some
patches need to be backported (PR45971 comes to my mind).

Reducing.

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> 2010-11-30 18:11:41 UTC ---
Created attachment 22583
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22583
A reduced testcase

[hjl@gnu-35 delta]$ /usr/gcc-4.5/bin/gcc -S -O3 testcase-min.cc
testcase-min.cc: In function ‘void testVectors() [with V =
Vc::Simple::Vector<short int>]’:
testcase-min.cc:157:27: internal compiler error: in
get_initial_def_for_induction, at tree-vect-loop.c:2431
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
[hjl@gnu-35 delta]$

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #4 from H.J. Lu <hjl.tools at gmail dot com> 2010-11-30 19:20:29 UTC ---
It is caused by revision 161951:

http://gcc.gnu.org/ml/gcc-cvs/2010-07/msg00305.html

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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 #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-01 10:01:32 UTC ---
Mine then.

[Bug tree-optimization/46723] [4.5/4.6 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to work|4.6.0                       |
            Summary|[4.5 Regression] internal   |[4.5/4.6 Regression]
                   |compiler error: in          |internal compiler error: in
                   |get_initial_def_for_inducti |get_initial_def_for_inducti
                   |on, at                      |on, at
                   |tree-vect-loop.c:2431       |tree-vect-loop.c:2431
      Known to fail|                            |4.6.0

--- Comment #6 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-01 15:32:55 UTC ---
C testcase that also ICEs on trunk:

short *m;
void test()
{
  short x = 128;
  unsigned int i;
  for (i = 0; i < 128; ++i, x = (unsigned short)x + 1)
    m[i] = x;
}

we are confused wrt the address arithmetic in

  D.2720_6 = (long unsigned int) i_20;
  D.2721_7 = D.2720_6 * 2;
  D.2722_8 = pretmp.6_18 + D.2721_7;
  *D.2722_8 = x_19;

and don't ICE with

short m[128];

which just gives us m[i_20] = x_19;

The access function that isn't a simple iv evolution is:

(short int) {(short unsigned int) x_50, +, 1}_1

At analysis time (in the not copied loop) we analyse it to

{128, +, 1}_1

instead as the loop bound is now non-constant after peeling a prologue:

<bb 3>:
  # x_19 = PHI <x_12(4), x_50(22)>
  # i_20 = PHI <i_9(4), i_52(22)>
  # ivtmp.14_5 = PHI <ivtmp.14_27(4), ivtmp.14_55(22)>
  D.2690_6 = (long unsigned int) i_20;
  D.2691_7 = D.2690_6 * 2;
  D.2692_8 = pretmp.6_18 + D.2691_7;
  *D.2692_8 = x_19;
  i_9 = i_20 + 1;
  x.1_10 = (short unsigned int) x_19;
  D.2694_11 = x.1_10 + 1;
  x_12 = (short int) D.2694_11;
  ivtmp.14_27 = ivtmp.14_5 - 1;
  if (ivtmp.14_27 != 0)
    goto <bb 4>;
  else
    goto <bb 19>;

in principle we can deal with that evolution just fine, but this
discrepancy analysis vs. transformation time doesn't look too good.

[Bug tree-optimization/46723] [4.5/4.6 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #7 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-01 15:37:09 UTC ---
Created attachment 22587
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22587
patch

Patch I'm testing.  Computes the induction PHI in the type of the evolution,
stripping conversions (sign-changing only for now) and re-applying them at
uses (by inserting another assignment that does the conversion in place for
the vectorized stmt).

[Bug tree-optimization/46723] [4.5/4.6 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[irar at il dot ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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

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

--- Comment #8 from Ira Rosen <irar at il dot ibm.com> 2010-12-02 08:15:16 UTC ---
*************** get_initial_def_for_induction (gimple iv
*** 2656,2667 ****
--- 2639,2664 ----
    loop_arg = PHI_ARG_DEF_FROM_EDGE (iv_phi, latch_e);

    access_fn = analyze_scalar_evolution (iv_loop, PHI_RESULT (iv_phi));
+   STRIP_NOPS (access_fn);
    gcc_assert (access_fn);

this makes gcc_assert meaningless: STRIP_NOPS will segfault for NULL access_fn
before the assert.

Thanks,
Ira

[Bug tree-optimization/46723] [4.5/4.6 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenther at suse dot de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #9 from rguenther at suse dot de <rguenther at suse dot de> 2010-12-02 09:03:35 UTC ---
On Thu, 2 Dec 2010, irar at il dot ibm.com wrote:

> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723
> 
> Ira Rosen <irar at il dot ibm.com> changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>                  CC|                            |irar at il dot ibm.com
> 
> --- Comment #8 from Ira Rosen <irar at il dot ibm.com> 2010-12-02 08:15:16 UTC ---
> *************** get_initial_def_for_induction (gimple iv
> *** 2656,2667 ****
> --- 2639,2664 ----
>     loop_arg = PHI_ARG_DEF_FROM_EDGE (iv_phi, latch_e);
> 
>     access_fn = analyze_scalar_evolution (iv_loop, PHI_RESULT (iv_phi));
> +   STRIP_NOPS (access_fn);
>     gcc_assert (access_fn);
> 
> this makes gcc_assert meaningless: STRIP_NOPS will segfault for NULL access_fn
> before the assert.

Ah, indeed.  Of course we should never assert here (if we do the analysis
has wrongly determined we can handle the induction), but of course we
can't fail to vectorize at this point either ...

Thanks,
Richard.

[Bug tree-optimization/46723] [4.5/4.6 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #10 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-02 16:23:26 UTC ---
Author: rguenth
Date: Thu Dec  2 16:23:20 2010
New Revision: 167377

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=167377
Log:
2010-12-02  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/46723
    * tree-vect-loop.c (get_initial_def_for_induction): Strip
    conversions from the induction evolution and apply it to
    the result instead.
    * tree-vect-stmts.c (vect_get_vec_def_for_operand): Handle
    assigns for induction defs.

    * gcc.dg/torture/pr46723.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/torture/pr46723.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-vect-loop.c
    trunk/gcc/tree-vect-stmts.c

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2
      Known to work|                            |4.6.0
            Summary|[4.5/4.6 Regression]        |[4.5 Regression] internal
                   |internal compiler error: in |compiler error: in
                   |get_initial_def_for_inducti |get_initial_def_for_inducti
                   |on, at                      |on, at
                   |tree-vect-loop.c:2431       |tree-vect-loop.c:2431
      Known to fail|4.6.0                       |
           Severity|critical                    |normal

--- Comment #11 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-02 16:27:26 UTC ---
Fixed on trunk sofar.

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.5.2                       |4.5.3

--- Comment #12 from Richard Guenther <rguenth at gcc dot gnu.org> 2010-12-16 13:02:42 UTC ---
GCC 4.5.2 is being released, adjusting target milestone.

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[doko at ubuntu dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

Matthias Klose <doko at ubuntu dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |doko at ubuntu dot com

--- Comment #13 from Matthias Klose <doko at ubuntu dot com> 2011-02-19 18:43:00 UTC ---
the patch fixes the ICE on the 4.5 branch, no regressions on a i686-linux-gnu
biarch build and test.

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

--- Comment #14 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-03-01 17:04:32 UTC ---
Author: rguenth
Date: Tue Mar  1 17:04:26 2011
New Revision: 170595

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=170595
Log:
2011-03-01  Richard Guenther  <rguenther@suse.de>

        Backport from mainline
        2011-03-01  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/47890
    * tree-vect-loop.c (get_initial_def_for_induction): Set
    related stmt properly.

    * gcc.dg/torture/pr47890.c: New testcase.

        2010-12-01  Richard Guenther  <rguenther@suse.de>

    PR tree-optimization/46723
    * tree-vect-loop.c (get_initial_def_for_induction): Strip
    conversions from the induction evolution and apply it to
    the result instead.
    * tree-vect-stmts.c (vect_get_vec_def_for_operand): Handle
    assigns for induction defs.

    * gcc.dg/torture/pr46723.c: New testcase.

Added:
    branches/gcc-4_5-branch/gcc/testsuite/gcc.dg/torture/pr46723.c
    branches/gcc-4_5-branch/gcc/testsuite/gcc.dg/torture/pr47890.c
Modified:
    branches/gcc-4_5-branch/gcc/ChangeLog
    branches/gcc-4_5-branch/gcc/testsuite/ChangeLog
    branches/gcc-4_5-branch/gcc/tree-vect-loop.c
    branches/gcc-4_5-branch/gcc/tree-vect-stmts.c

[Bug tree-optimization/46723] [4.5 Regression] internal compiler error: in get_initial_def_for_induction, at tree-vect-loop.c:2431

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46723

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

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

--- Comment #15 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-03-01 17:04:58 UTC ---
Fixed.