[Bug ada/28323] New: segfault in ada_is_unconstrained_packed_array_type

[Bug ada/28323] New: segfault in ada_is_unconstrained_packed_array_type

[vries at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=28323

            Bug ID: 28323
           Summary: segfault in ada_is_unconstrained_packed_array_type
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: ada
          Assignee: unassigned at sourceware dot org
          Reporter: vries at gcc dot gnu.org
  Target Milestone: ---

With test-case rtf=gdb.ada/mi_dyn_arr.exp I run into a segfault on openSUSE
Leap 42.3 with gcc-4.8:
...
(gdb) ^M
Expecting: ^(-var-create bt \* bt[^M
]+)?(\^done,name="bt",numchild="3",.*[^M
]+[(]gdb[)] ^M
[ ]*)
-var-create bt * bt^M
ERROR: Process no longer exists
UNRESOLVED: gdb.ada/mi_dyn_arr.exp: create bt varobj
ERROR: Couldn't send -var-update 1 * to GDB.
...

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug ada/28323] segfault in ada_is_unconstrained_packed_array_type

[vries at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=28323

--- Comment #1 from Tom de Vries <vries at gcc dot gnu.org> ---
Reproduce:
...
$ cat gdb.in
interpreter-exec mi "100-gdb-set height 0"
interpreter-exec mi "101-gdb-set width 0"
show mi-async
interpreter-exec mi "104-environment-directory -r"
interpreter-exec mi "105-environment-directory
/home/vries/gdb_versions/devel/src/gdb/testsuite/gdb.ada"
interpreter-exec mi "105-file-exec-and-symbols
/home/vries/gdb_versions/devel/foo"
interpreter-exec mi "200-break-insert --qualified -t main"
interpreter-exec mi "220-exec-run"
interpreter-exec mi "-break-insert -t foo.adb:22"
interpreter-exec mi "-exec-continue"
interpreter-exec mi "-var-create bt * bt"
interpreter-exec mi "-var-update 1 *"
$ gdb.sh -q -batch -x gdb.in
...
[ note: with foo copied out of build tree, not relevant to reproducing though ]

Problem is here:
...
Thread 1 "gdb" received signal SIGSEGV, Segmentation fault.
0x0000000000450861 in ada_is_unconstrained_packed_array_type (type=0x25163f0)
    at /home/vries/gdb_versions/devel/src/gdb/ada-lang.c:1978
1978          return TYPE_FIELD_BITSIZE (type, 0) > 0;
(gdb) p *type
$2 = {pointer_type = 0x2ae4e20, reference_type = 0x0, rvalue_reference_type =
0x0, 
  chain = 0x25163f0, align_log2 = 0, m_instance_flags = 0, length = 0,
main_type = 0x2ae4900}
(gdb) p type.main_type->flds_bnds.fields
$4 = (field *) 0x0
...

Type in more detail:
...
(gdb) p *type.main_type
$3 = {code = TYPE_CODE_TYPEDEF, m_flag_unsigned = 0, m_flag_nosign = 0,
m_flag_stub = 0, 
  m_flag_target_stub = 1, m_flag_prototyped = 0, m_flag_varargs = 0,
m_flag_vector = 0, 
  m_flag_stub_supported = 0, m_flag_gnu_ifunc = 0, m_flag_fixed_instance = 0, 
  m_flag_objfile_owned = 1, m_flag_endianity_not_default = 0,
m_flag_declared_class = 0, 
  m_flag_flag_enum = 0, type_specific_field = TYPE_SPECIFIC_GNAT_STUFF, nfields
= 0, 
  name = 0x7fffec9733a2 "pck__bounded___XUA", m_owner = {objfile = 0x1b4e6f0,
gdbarch = 
    0x1b4e6f0}, target_type = 0x2ae4b80, flds_bnds = {fields = 0x0, bounds =
0x0, 
    complex_type = 0x0}, type_specific = {cplus_stuff = 0xf2e9c8
<gnat_aux_default>, 
    gnat_stuff = 0xf2e9c8 <gnat_aux_default>, floatformat = 0xf2e9c8
<gnat_aux_default>, 
    func_stuff = 0xf2e9c8 <gnat_aux_default>, self_type = 0xf2e9c8
<gnat_aux_default>, 
    fixed_point_info = 0xf2e9c8 <gnat_aux_default>, int_stuff = {bit_size =
59848, 
      bit_offset = 242}}, dyn_prop_list = 0x0}
...

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug ada/28323] segfault in ada_is_unconstrained_packed_array_type

[vries at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=28323

--- Comment #2 from Tom de Vries <vries at gcc dot gnu.org> ---
This seems to fix it:
...
$ git diff
diff --git a/gdb/ada-lang.c b/gdb/ada-lang.c
index 6680a4fd657..b1ed66c12c4 100644
--- a/gdb/ada-lang.c
+++ b/gdb/ada-lang.c
@@ -1974,6 +1974,8 @@ ada_is_unconstrained_packed_array_type (struct type
*type)
       /* The structure's first field is a pointer to an array, so this
         fetches the array type.  */
       type = TYPE_TARGET_TYPE (type->field (0).type ());
+      if (type->code () == TYPE_CODE_TYPEDEF)
+       type = ada_typedef_target_type (type);
       /* Now we can see if the array elements are packed.  */
       return TYPE_FIELD_BITSIZE (type, 0) > 0;
     }
...

But I have no idea whether the fix is correct.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug ada/28323] segfault in ada_is_unconstrained_packed_array_type

[vries at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=28323

Tom de Vries <vries at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |brobecke at sourceware dot org,
                   |                            |tromey at sourceware dot org

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug ada/28323] segfault in ada_is_unconstrained_packed_array_type

[vries at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=28323

--- Comment #3 from Tom de Vries <vries at gcc dot gnu.org> ---
https://sourceware.org/pipermail/gdb-patches/2021-December/184241.html

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug ada/28323] segfault in ada_is_unconstrained_packed_array_type

[vries at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=28323

Tom de Vries <vries at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|NEW                         |RESOLVED
   Target Milestone|---                         |12.1

--- Comment #4 from Tom de Vries <vries at gcc dot gnu.org> ---
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=af5300fe24bf2f4e032d639a5396a16c1762b77b

-- 
You are receiving this mail because:
You are on the CC list for the bug.