[Bug fortran/110076] New: ICE on mutually recursive derived types.

[Bug fortran/110076] New: ICE on mutually recursive derived types.

[srm.1708 at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110076

            Bug ID: 110076
           Summary: ICE on mutually recursive derived types.
           Product: gcc
           Version: 12.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: srm.1708 at gmail dot com
  Target Milestone: ---

Minimal repro is:

$ cat main.f90
module foo
    type bar
        type(baz), allocatable :: a
    end type

    type baz
        type(bar), allocatable :: b
    end type
end module
$ gfortran -c main.f90
gfortran: internal compiler error: Segmentation fault: 11 signal terminated
program f951
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
See <https://gcc.gnu.org/bugs/> for instructions.

Removing the allocatable statements or making it singly recursive (i.e.
removing the baz type and giving bar a field of type(bar)) avoids the ICE.

This is potentially a duplicate of 94599 or 104048, but neither of those is
exhibited _only_ when the types are mutually recursive so I thought I should
open this up any way.

I've tested this with two different versions of GFortran:

$ gfortran -v
Using built-in specs.
COLLECT_GCC=/nix/store/gg2pvn8xifrwbaa09msl6k9g921j1595-gfortran-12.2.0/bin/gfortran
COLLECT_LTO_WRAPPER=/nix/store/gg2pvn8xifrwbaa09msl6k9g921j1595-gfortran-12.2.0/libexec/gcc/aarch64-apple-darwin/12.2.0/lto-wrapper
Target: aarch64-apple-darwin
Configured with:
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 12.2.0 (GCC)

$ gfortran -v
Using built-in specs.
COLLECT_GCC=gfortran
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/10/lto-wrapper
OFFLOAD_TARGET_NAMES=nvptx-none:amdgcn-amdhsa:hsa
OFFLOAD_TARGET_DEFAULT=1
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Debian 10.2.1-6'
--with-bugurl=file:///usr/share/doc/gcc-10/README.Bugs
--enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --prefix=/usr
--with-gcc-major-version-only --program-suffix=-10
--program-prefix=x86_64-linux-gnu- --enable-shared --enable-linker-build-id
--libexecdir=/usr/lib --without-included-gettext --enable-threads=posix
--libdir=/usr/lib --enable-nls --enable-bootstrap --enable-clocale=gnu
--enable-libstdcxx-debug --enable-libstdcxx-time=yes
--with-default-libstdcxx-abi=new --enable-gnu-unique-object
--disable-vtable-verify --enable-plugin --enable-default-pie --with-system-zlib
--enable-libphobos-checking=release --with-target-system-zlib=auto
--enable-objc-gc=auto --enable-multiarch --disable-werror --with-arch-32=i686
--with-abi=m64 --with-multilib-list=m32,m64,mx32 --enable-multilib
--with-tune=generic
--enable-offload-targets=nvptx-none=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-gcn/usr,hsa
--without-cuda-driver --enable-checking=release --build=x86_64-linux-gnu
--host=x86_64-linux-gnu --target=x86_64-linux-gnu
--with-build-config=bootstrap-lto-lean --enable-link-mutex
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 10.2.1 20210110 (Debian 10.2.1-6)

[Bug fortran/110076] ICE on mutually recursive derived types.

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110076

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-06-01
     Ever confirmed|0                           |1

--- Comment #1 from anlauf at gcc dot gnu.org ---
Confirmed.  I believe I've seen another duplicate where running under gdb
shows that we enter an infinite loop in gfc_get_derived_type:

(gdb) l 2731,2753
2731      /* Go through the derived type components, building them as
2732         necessary. The reason for doing this now is that it is
2733         possible to recurse back to this derived type through a
2734         pointer component (PR24092). If this happens, the fields
2735         will be built and so we can return the type.  */
2736      for (c = derived->components; c; c = c->next)
2737        {
2738          bool same_alloc_type = c->attr.allocatable
2739                                 && derived == c->ts.u.derived;
2740
2741          if (c->ts.type == BT_UNION && c->ts.u.derived->backend_decl ==
NULL)
2742            c->ts.u.derived->backend_decl = gfc_get_union_type
(c->ts.u.derived);
2743
2744          if (c->ts.type != BT_DERIVED && c->ts.type != BT_CLASS)
2745            continue;
2746
2747          if ((!c->attr.pointer && !c->attr.proc_pointer
2748              && !same_alloc_type)
2749              || c->ts.u.derived->backend_decl == NULL)
2750            {
2751              int local_codim = c->attr.codimension ? c->as->corank:
codimen;
2752              c->ts.u.derived->backend_decl = gfc_get_derived_type
(c->ts.u.derived,
2753                                                                   
local_codim);