[Bug target/96456] New: [10 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[Bug target/96456] New: [10 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[doko at debian dot org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

            Bug ID: 96456
           Summary: [10 Regression] ICE in expand_insn, at optabs.c:7511
                    on s390x-linux-gnu
           Product: gcc
           Version: 10.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: doko at debian dot org
  Target Milestone: ---

seen with the gcc-10 branch 20200725 on s390x-linux-gnu. worked around by
dropping to -O2, or either removing the -ffast-math or -fno-finite-math-only
flags. Didn't check yet trunk.

$ cat aops.i
int b, c, d;
double *e;
int f() {
  double *a = a;
  int g = d, f = c, h = b;
  if (__builtin_expect(f, 0))
    for (; g < h; g++)
      e[g] = (int)(a[g] >= 0.0 ? g + 0.99999999 : a[g]);
  return 0;
}

$ s390x-linux-gnu-gcc -c -O3 -std=gnu99 -ffast-math -fno-finite-math-only
-fstack-protector-strong -Wall -fstack-clash-protection aops.i 
during RTL pass: expand
aops.i: In function ‘f’:
aops.i:8:12: internal compiler error: in expand_insn, at optabs.c:7511
    8 |       e[g] = (int)(a[g] >= 0.0 ? g + 0.99999999 : a[g]);
      |       ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Please submit a full bug report,
with preprocessed source if appropriate.

[Bug target/96456] [10/11 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|                            |10.2.0
            Summary|[10 Regression] ICE in      |[10/11 Regression] ICE in
                   |expand_insn, at             |expand_insn, at
                   |optabs.c:7511 on            |optabs.c:7511 on
                   |s390x-linux-gnu             |s390x-linux-gnu
   Target Milestone|---                         |10.3

[Bug target/96456] [10/11 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[krebbel at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

--- Comment #1 from Andreas Krebbel <krebbel at gcc dot gnu.org> ---
Confirmed for current gcc 10 branch. Does not appear to happen on trunk though.
I'll have a look.

[Bug target/96456] [10/11 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

--- Comment #2 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Andreas Krebbel <krebbel@gcc.gnu.org>:

https://gcc.gnu.org/g:1d17c38ac1d1d916abbce14e6b2e0d61517c6e9b

commit r11-2661-g1d17c38ac1d1d916abbce14e6b2e0d61517c6e9b
Author: Andreas Krebbel <krebbel@linux.ibm.com>
Date:   Wed Aug 12 08:02:34 2020 +0200

    IBM Z: Fix PR96456

    The testcase failed because our backend refuses to generate vector
    compare instructions for signaling operators with -fno-trapping-math
    -fno-finite-math-only.

    gcc/ChangeLog:

            PR target/96456
            * config/s390/s390.h (TARGET_NONSIGNALING_VECTOR_COMPARE_OK): New
            macro.
            * config/s390/vector.md (vcond_comparison_operator): Use new macro
            for the check.

    gcc/testsuite/ChangeLog:

            PR target/96456
            * gcc.target/s390/pr96456.c: New test.

[Bug target/96456] [10/11 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

--- Comment #3 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-10 branch has been updated by Andreas Krebbel
<krebbel@gcc.gnu.org>:

https://gcc.gnu.org/g:537f9f4e5b12f4ef5ca363f89466c0470f3482bf

commit r10-8856-g537f9f4e5b12f4ef5ca363f89466c0470f3482bf
Author: Andreas Krebbel <krebbel@linux.ibm.com>
Date:   Wed Aug 12 08:02:34 2020 +0200

    IBM Z: Fix PR96456

    The testcase failed because our backend refuses to generate vector
    compare instructions for signaling operators with -fno-trapping-math
    -fno-finite-math-only.

    gcc/ChangeLog:

            PR target/96456
            * config/s390/s390.h (TARGET_NONSIGNALING_VECTOR_COMPARE_OK): New
            macro.
            * config/s390/vector.md (vcond_comparison_operator): Use new macro
            for the check.

    gcc/testsuite/ChangeLog:

            PR target/96456
            * gcc.target/s390/pr96456.c: New test.

    (cherry picked from commit 1d17c38ac1d1d916abbce14e6b2e0d61517c6e9b)

[Bug target/96456] [10/11 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[krebbel at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

Andreas Krebbel <krebbel at gcc dot gnu.org> changed:

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

--- Comment #4 from Andreas Krebbel <krebbel at gcc dot gnu.org> ---
Fixed for trunk and GCC 10

[Bug target/96456] [10/11 Regression] ICE in expand_insn, at optabs.c:7511 on s390x-linux-gnu

[krebbel at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96456

Andreas Krebbel <krebbel at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #5 from Andreas Krebbel <krebbel at gcc dot gnu.org> ---
closing