1 # Copyright 2002-2024 Free Software Foundation, Inc.
3 # This program is free software; you can redistribute it and/or modify
4 # it under the terms of the GNU General Public License as published by
5 # the Free Software Foundation; either version 3 of the License, or
6 # (at your option) any later version.
8 # This program is distributed in the hope that it will be useful,
9 # but WITHOUT ANY WARRANTY; without even the implied warranty of
10 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11 # GNU General Public License for more details.
13 # You should have received a copy of the GNU General Public License
14 # along with this program. If not, see <http://www.gnu.org/licenses/>.
17 require allow_cplus_tests
19 standard_testfile hang1.cc hang2.cc hang3.cc
23 if {[prepare_for_testing "failed to prepare" $testfile \
24 [list $srcfile $srcfile2 $srcfile3] {debug c++}]} {
28 # As of May 1, 2002, GDB hangs trying to read the debug info for the
29 # `hang2.o' compilation unit from the executable `hang', when compiled
30 # by g++ 2.96 with STABS debugging info. Here's what's going on, as
33 # The definition of `struct A' in `hang.H' refers to `struct B' as an
34 # incomplete type. The stabs declare type number (1,3) to be a cross-
35 # reference type, `xsB:'.
37 # The definition of `struct C' contains a nested definition for
38 # `struct B' --- or more properly, `struct C::B'. However, the stabs
39 # fail to qualify the structure tag: it just looks like a definition
40 # for `struct B'. I think this is a compiler bug, but perhaps GCC
41 # doesn't emit qualified names for a reason.
43 # `hang.H' gets #included by both `hang1.C' and `hang2.C'. So the
44 # stabs for `struct A', the incomplete `struct B', and `struct C'
45 # appear in both hang1.o's and hang2.o's stabs.
47 # When those two files are linked together, since hang2.o appears
48 # later in the command line, its #inclusion of `hang.H' gets replaced
49 # with an N_EXCL stab, referring back to hang1.o's stabs for the
52 # When GDB builds psymtabs for the executable hang, it notes that
53 # hang2.o's stabs contain an N_EXCL referring to a header that appears
54 # in full in hang1.o's stabs. So hang2.o's psymtab lists a dependency
55 # on hang1.o's psymtab.
57 # When the user types the command `print var_in_b', GDB scans the
58 # psymtabs for a symbol by that name, and decides to read full symbols
61 # Since `hang2.o''s psymtab lists `hang1.o' as a dependency, GDB first
62 # reads `hang1.o''s symbols. When GDB sees `(1,3)=xsB:', it creates a
63 # type object for `struct B', sets its TYPE_STUB flag, and records it
64 # as type number `(1,3)'.
66 # When GDB finds the definition of `struct C::B', since the stabs
67 # don't indicate that the type is nested within C, it treats it as
68 # a definition of `struct B'.
70 # When GDB is finished reading `hang1.o''s symbols, it calls
71 # `cleanup_undefined_types'. This function mistakes the definition of
72 # `struct C::B' for a definition for `struct B', and overwrites the
73 # incomplete type object for the real `struct B', using `memcpy'. Now
74 # stabs type number `(1,3)' refers to this (incorrect) complete type.
75 # Furthermore, the `memcpy' simply copies the original's `cv_type'
76 # field to the target, giving the target a corrupt `cv_type' ring: the
77 # chain does not point back to the target type.
79 # Having satisfied `hang2.o''s psymtab's dependencies, GDB begins to
80 # read `hang2.o''s symbols. These contain the true definition for
81 # `struct B', which refers to type number `(1,3)' as the type it's
82 # defining. GDB looks up type `(1,3)', and finds the (incorrect)
83 # complete type established by the call to `cleanup_undefined_types'
84 # above. However, it doesn't notice that the type is already defined,
85 # and passes it to `read_struct_type', which then writes the new
86 # definition's size, field list, etc. into the type object which
87 # already has those fields initialized. Adding insult to injury,
88 # `read_struct_type' then calls `finish_cv_type'; since the `memcpy'
89 # in `cleanup_undefined_types' corrupted the target type's `cv_type'
90 # ring, `finish_cv_type' enters an infinite loop.
92 # This checks that GDB recognizes when a structure is about to be
93 # overwritten, and refuses, with a complaint.
94 gdb_test "print var_in_b" " = 1729" "doesn't overwrite struct type"
96 # This checks that cleanup_undefined_types doesn't create corrupt
97 # cv_type chains. Note that var_in_hang3 does need to be declared in
98 # a separate compilation unit, whose psymtab depends on hang1.o's
99 # psymtab. Otherwise, GDB won't call cleanup_undefined_types (as it
100 # finishes hang1.o's symbols) before it calls make_cv_type (while
101 # reading hang3.o's symbols).
103 # The bug only happens when you compile with -gstabs+; Otherwise, GCC
104 # won't include the `const' qualifier on `const_B_ptr' in `hang3.o''s
105 # STABS, so GDB won't try to create a const variant of the smashed
106 # struct type, and get caught by the corrupted cv_type chain.
107 gdb_test "print var_in_hang3" " = 42" "doesn't corrupt cv_type chain"