[Bug fortran/66562] New: ICE with gfortran-5.1.0

[Bug fortran/66562] New: ICE with gfortran-5.1.0

[Daniel.Kokron at nasa dot gov]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66562

            Bug ID: 66562
           Summary: ICE with gfortran-5.1.0
           Product: gcc
           Version: 5.1.0
            Status: UNCONFIRMED
          Severity: blocker
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: Daniel.Kokron at nasa dot gov
  Target Milestone: ---

Freshly compiled gcc-5.1.0 using gmp-6.0.0, mpfr-3.1.2 and  mpc-1.0.3 give an
ICE.  The system is SLES11 sp3.  All of the above were compiled using gcc (SUSE
Linux) 4.3.4 [gcc-4_3-branch revision 152973

ldd
/nobackupnfs2/xxxx/Projects/GCC/gcc-5.1.0/install/libexec/gcc/x86_64-unknown-linux-gnu/5.1.0/f951
        linux-vdso.so.1 =>  (0x00007ffe3fffe000)
        libmpc.so.3 => /u/xxxx/play/MPC/mpc-1.0.3/install/lib/libmpc.so.3
(0x00007ffe3fde5000)
        libmpfr.so.4 => /u/xxxx/play/MPFR/mpfr-3.1.2/install/lib/libmpfr.so.4
(0x00007ffe3fb87000)
        libgmp.so.10 => /u/xxxx/play/GMP/gmp-6.0.0/install/lib/libgmp.so.10
(0x00007ffe3f910000)
        libdl.so.2 => /lib64/libdl.so.2 (0x00007ffe3f6ca000)
        libm.so.6 => /lib64/libm.so.6 (0x00007ffe3f450000)
        libc.so.6 => /lib64/libc.so.6 (0x00007ffe3f0d4000)
        /lib64/ld-linux-x86-64.so.2 (0x0000555555554000)

cat StringClass.f03
module StringClass
! gfortran -fPIC -Wall -ffree-form -O0 -m64 -c StringClass.f03
  implicit none
  private

  public :: String

  type :: String
    private
    integer, private              :: intMaxLength = 0
    character(len=:), allocatable :: strCharacters
  contains
    procedure, public,  pass(this) :: Countem
  end type String


contains


  function Countem (this, dmy_strDelims, dmy_blnIgnoreCase) result (intCount)
    implicit none
    class(String),               intent(inout) :: this
    character (len=*), optional, intent (in)   :: dmy_strDelims
    logical (kind=1),  optional, intent (in)   :: dmy_blnIgnoreCase
    integer (kind=4) :: intCount
    integer (kind=4) :: intScanPos

    intScanPos = SCAN (this%strCharacters(1:), dmy_strDelims)
  end function Countem


end module StringClass

which gfortran
/nobackupnfs2/xxxx/Projects/GCC/gcc-5.1.0/install/bin/gfortran

661> gfortran -fPIC -cpp -Wall -ffree-form -O2 -m64 -c StringClass.f03
f951: internal compiler error: Segmentation fault
0xa7af5f crash_signal
        ../.././gcc/toplev.c:383
0x6389fe gfc_copy_shape(__mpz_struct (*) [1], int)
        ../.././gcc/fortran/expr.c:708
0x638aa5 gfc_copy_expr(gfc_expr*)
        ../.././gcc/fortran/expr.c:393
0x6926f7 gfc_resolve_substring_charlen(gfc_expr*)
        ../.././gcc/fortran/resolve.c:4580
0x6900e5 gfc_resolve_expr(gfc_expr*)
        ../.././gcc/fortran/resolve.c:6260
0x6928f3 resolve_actual_arglist
        ../.././gcc/fortran/resolve.c:1930
0x690019 resolve_function
        ../.././gcc/fortran/resolve.c:2947
0x690019 gfc_resolve_expr(gfc_expr*)
        ../.././gcc/fortran/resolve.c:6254
0x696a4c gfc_resolve_code(gfc_code*, gfc_namespace*)
        ../.././gcc/fortran/resolve.c:10095
0x6993e2 resolve_codes
        ../.././gcc/fortran/resolve.c:15055
0x6992e7 resolve_codes
        ../.././gcc/fortran/resolve.c:15040
0x6994c2 gfc_resolve(gfc_namespace*)
        ../.././gcc/fortran/resolve.c:15083
0x684df6 gfc_parse_file()
        ../.././gcc/fortran/parse.c:5477
0x6c43f5 gfc_be_parse_file
        ../.././gcc/fortran/f95-lang.c:228
Please submit a full bug report,
with preprocessed source if appropriate.

[Bug fortran/66562] ICE with gfortran-5.1.0

[kargl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66562

kargl at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kargl at gcc dot gnu.org
           Severity|blocker                     |normal

--- Comment #1 from kargl at gcc dot gnu.org ---
(In reply to Dan Kokron from comment #0)

> 
>     intScanPos = SCAN (this%strCharacters(1:), dmy_strDelims)

Workaround is to not use an array section here.

PS: Fortran bugs are never have an importance of "blocker" unless
the bug breaks bootstrap.

[Bug fortran/66562] ICE with gfortran-5.1.0

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66562

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

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

--- Comment #2 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
I think it is a duplicate of pr63494.

*** This bug has been marked as a duplicate of bug 63494 ***

[Bug fortran/66562] ICE with gfortran-5.1.0

[Daniel.Kokron at nasa dot gov]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66562

--- Comment #3 from Dan Kokron <Daniel.Kokron at nasa dot gov> ---
@karlg  The original code accessed the array with in a loop using i:, so the
proposed workaround doesn't help my case.

PS: Thanks for the advise.  Newbie mistake.