[Bug tree-optimization/107365] New: ICE in verify_range, at value-range.cc:726

[Bug tree-optimization/107365] New: ICE in verify_range, at value-range.cc:726

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107365

            Bug ID: 107365
           Summary: ICE in verify_range, at value-range.cc:726
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: marxin at gcc dot gnu.org
                CC: aldyh at gcc dot gnu.org
  Target Milestone: ---
              Host: x86_64-linux-gnu
            Target: vax-linux-gnu

Happens for a cross compiler:

$ ./xgcc -B./ -B/home/marxin/bin/gcc/vax-linux-gnu/bin/ -isystem
/home/marxin/bin/gcc/vax-linux-gnu/include -isystem
/home/marxin/bin/gcc/vax-linux-gnu/sys-include -L/dev/shm/objdir2/gcc/../ld 
-xc++ -nostdinc /dev/null -S -o /dev/null
-fself-test=/home/marxin/Programming/gcc/gcc/testsuite/selftests
In function ‘test_fn’:
cc1plus: internal compiler error: in verify_range, at value-range.cc:726
0x18dbdc5 frange::verify_range()
        /home/marxin/Programming/gcc/gcc/value-range.cc:726
0x18da448 frange::set(tree_node*, real_value const&, real_value const&,
value_range_kind)
        /home/marxin/Programming/gcc/gcc/value-range.cc:366
0x18f17fe frange::frange(tree_node*, real_value const&, real_value const&,
value_range_kind)
        /home/marxin/Programming/gcc/gcc/value-range.h:1080
0x18ed069 frange_float
        /home/marxin/Programming/gcc/gcc/value-range.cc:3692
0x18f021c range_tests_floats
        /home/marxin/Programming/gcc/gcc/value-range.cc:3959
0x18f11c9 selftest::range_tests()
        /home/marxin/Programming/gcc/gcc/value-range.cc:4052
0x20549b7 test_ranges
        /home/marxin/Programming/gcc/gcc/function-tests.cc:584
0x20549b7 selftest::function_tests_cc_tests()
        /home/marxin/Programming/gcc/gcc/function-tests.cc:680
0x21b488c selftest::run_tests()
        /home/marxin/Programming/gcc/gcc/selftest-run-tests.cc:107
0x146880e toplev::run_self_tests()
        /home/marxin/Programming/gcc/gcc/toplev.cc:2184
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
Please include the complete backtrace with any bug report.
See <https://gcc.gnu.org/bugs/> for instructions.

[Bug tree-optimization/107365] ICE in verify_range, at value-range.cc:726

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107365

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2022-10-23
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
                 CC|                            |amacleod at redhat dot com

[Bug tree-optimization/107365] ICE in verify_range, at value-range.cc:726

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107365

--- Comment #1 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
Ok, this is getting ridiculous.  I'm tired of these weird finite-math-only
combinations in Vax and rx-elf.  I think we should just test -ffinite-math-only
and -fno-finite-math-only in the self tests for all architectures.  Heck, we
should probably check -fno-finite-math-only + !HONOR_NANS (float_type_mode).

[Bug tree-optimization/107365] ICE in verify_range, at value-range.cc:726

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107365

--- Comment #2 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
Created attachment 53761
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=53761&action=edit
untested

[Bug tree-optimization/107365] ICE in verify_range, at value-range.cc:726

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107365

--- Comment #3 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Aldy Hernandez <aldyh@gcc.gnu.org>:

https://gcc.gnu.org/g:1e9d9ed095df3d064cf9d91d46f3e5426c2a05a7

commit r13-3453-g1e9d9ed095df3d064cf9d91d46f3e5426c2a05a7
Author: Aldy Hernandez <aldyh@redhat.com>
Date:   Sun Oct 23 16:51:17 2022 +0200

    Check HONOR_NANS instead of flag_finite_math_only in frange:verify_range.

    [Jakub and other FP experts, would this be OK, or am I missing
    something?]

    Vax does not seem to have !flag_finite_math_only, but float_type_node
    does not HONOR_NANS.  The check in frange::verify_range dependend on
    flag_finite_math_only, which is technically not correct since
    frange::set_varying() checks HONOR_NANS instead of
    flag_finite_math_only.

    I'm actually getting tired of flag_finite_math_only and
    !flag_finite_math_only discrepancies in the selftests (Vax and rx-elf
    come to mind).  I think we should just test both alternatives in the
    selftests as in this patch.

    We could also check flag_finite_math_only=0 with a float_type_node
    that does not HONOR_NANs, but I have no idea how to twiddle
    FLOAT_MODE_FORMAT temporarily, and that may be over thinking it.

            PR tree-optimization/107365

    gcc/ChangeLog:

            * value-range.cc (frange::verify_range): Predicate NAN check in
            VARYING range on HONOR_NANS instead of flag_finite_math_only.
            (range_tests_floats): Same.
            (range_tests_floats_various): New.
            (range_tests): Call range_tests_floats_various.

[Bug tree-optimization/107365] ICE in verify_range, at value-range.cc:726

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107365

Aldy Hernandez <aldyh at gcc dot gnu.org> changed:

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

--- Comment #4 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
fixed