* [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug
@ 2022-01-26 19:55 Simon Marchi
2022-01-26 23:46 ` Lancelot SIX
2022-01-27 20:45 ` Tom Tromey
0 siblings, 2 replies; 5+ messages in thread
From: Simon Marchi @ 2022-01-26 19:55 UTC (permalink / raw)
To: gdb-patches
New in v2:
- formatting fixes in the .exp file
- add producer (GCC 11) check, update test file accordingly to contain
a producer string
- issue complaint when applying the fixup
g++ 11.1.0 has a bug where it will emit a negative
DW_AT_data_member_location in some cases:
$ cat test.cpp
#include <memory>
int
main()
{
std::unique_ptr<int> ptr;
}
$ g++ -g test.cpp
$ llvm-dwarfdump -F a.out
...
0x00000964: DW_TAG_member
DW_AT_name [DW_FORM_strp] ("_M_head_impl")
DW_AT_decl_file [DW_FORM_data1] ("/usr/include/c++/11.1.0/tuple")
DW_AT_decl_line [DW_FORM_data1] (125)
DW_AT_decl_column [DW_FORM_data1] (0x27)
DW_AT_type [DW_FORM_ref4] (0x0000067a "default_delete<int>")
DW_AT_data_member_location [DW_FORM_sdata] (-1)
...
This leads to a GDB crash (when built with ASan, otherwise probably
garbage results), since it tries to read just before (to the left, in
ASan speak) of the value's buffer:
==888645==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x6020000c52af at pc 0x7f711b239f4b bp 0x7fff356bd470 sp 0x7fff356bcc18
READ of size 1 at 0x6020000c52af thread T0
#0 0x7f711b239f4a in __interceptor_memcpy /build/gcc/src/gcc/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:827
#1 0x555c4977efa1 in value_contents_copy_raw /home/simark/src/binutils-gdb/gdb/value.c:1347
#2 0x555c497909cd in value_primitive_field(value*, long, int, type*) /home/simark/src/binutils-gdb/gdb/value.c:3126
#3 0x555c478f2eaa in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:333
#4 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#5 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#6 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#7 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#8 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#9 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#10 0x555c4760d45f in c_value_print_struct /home/simark/src/binutils-gdb/gdb/c-valprint.c:383
#11 0x555c4760df4c in c_value_print_inner(value*, ui_file*, int, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:438
#12 0x555c483ff9a7 in language_defn::value_print_inner(value*, ui_file*, int, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:632
#13 0x555c49758b68 in do_val_print /home/simark/src/binutils-gdb/gdb/valprint.c:1048
#14 0x555c49759b17 in common_val_print(value*, ui_file*, int, value_print_options const*, language_defn const*) /home/simark/src/binutils-gdb/gdb/valprint.c:1151
#15 0x555c478f2fcb in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:335
#16 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#17 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#18 0x555c4760d45f in c_value_print_struct /home/simark/src/binutils-gdb/gdb/c-valprint.c:383
#19 0x555c4760df4c in c_value_print_inner(value*, ui_file*, int, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:438
#20 0x555c483ff9a7 in language_defn::value_print_inner(value*, ui_file*, int, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:632
#21 0x555c49758b68 in do_val_print /home/simark/src/binutils-gdb/gdb/valprint.c:1048
#22 0x555c49759b17 in common_val_print(value*, ui_file*, int, value_print_options const*, language_defn const*) /home/simark/src/binutils-gdb/gdb/valprint.c:1151
#23 0x555c478f2fcb in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:335
#24 0x555c4760d45f in c_value_print_struct /home/simark/src/binutils-gdb/gdb/c-valprint.c:383
#25 0x555c4760df4c in c_value_print_inner(value*, ui_file*, int, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:438
#26 0x555c483ff9a7 in language_defn::value_print_inner(value*, ui_file*, int, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:632
#27 0x555c49758b68 in do_val_print /home/simark/src/binutils-gdb/gdb/valprint.c:1048
#28 0x555c49759b17 in common_val_print(value*, ui_file*, int, value_print_options const*, language_defn const*) /home/simark/src/binutils-gdb/gdb/valprint.c:1151
#29 0x555c4760f04c in c_value_print(value*, ui_file*, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:587
#30 0x555c483ff954 in language_defn::value_print(value*, ui_file*, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:614
#31 0x555c49759f61 in value_print(value*, ui_file*, value_print_options const*) /home/simark/src/binutils-gdb/gdb/valprint.c:1189
#32 0x555c48950f70 in print_formatted /home/simark/src/binutils-gdb/gdb/printcmd.c:337
#33 0x555c48958eda in print_value(value*, value_print_options const&) /home/simark/src/binutils-gdb/gdb/printcmd.c:1258
#34 0x555c48959891 in print_command_1 /home/simark/src/binutils-gdb/gdb/printcmd.c:1367
#35 0x555c4895a3df in print_command /home/simark/src/binutils-gdb/gdb/printcmd.c:1458
#36 0x555c4767f974 in do_simple_func /home/simark/src/binutils-gdb/gdb/cli/cli-decode.c:97
#37 0x555c47692e25 in cmd_func(cmd_list_element*, char const*, int) /home/simark/src/binutils-gdb/gdb/cli/cli-decode.c:2475
#38 0x555c4936107e in execute_command(char const*, int) /home/simark/src/binutils-gdb/gdb/top.c:670
#39 0x555c485f1bff in catch_command_errors /home/simark/src/binutils-gdb/gdb/main.c:523
#40 0x555c485f249c in execute_cmdargs /home/simark/src/binutils-gdb/gdb/main.c:618
#41 0x555c485f6677 in captured_main_1 /home/simark/src/binutils-gdb/gdb/main.c:1317
#42 0x555c485f6c83 in captured_main /home/simark/src/binutils-gdb/gdb/main.c:1338
#43 0x555c485f6d65 in gdb_main(captured_main_args*) /home/simark/src/binutils-gdb/gdb/main.c:1363
#44 0x555c46e41ba8 in main /home/simark/src/binutils-gdb/gdb/gdb.c:32
#45 0x7f71198bcb24 in __libc_start_main (/usr/lib/libc.so.6+0x27b24)
#46 0x555c46e4197d in _start (/home/simark/build/binutils-gdb-one-target/gdb/gdb+0x77f197d)
0x6020000c52af is located 1 bytes to the left of 8-byte region [0x6020000c52b0,0x6020000c52b8)
allocated by thread T0 here:
#0 0x7f711b2b7459 in __interceptor_calloc /build/gcc/src/gcc/libsanitizer/asan/asan_malloc_linux.cpp:154
#1 0x555c470acdc9 in xcalloc /home/simark/src/binutils-gdb/gdb/alloc.c:100
#2 0x555c49b775cd in xzalloc(unsigned long) /home/simark/src/binutils-gdb/gdbsupport/common-utils.cc:29
#3 0x555c4977bdeb in allocate_value_contents /home/simark/src/binutils-gdb/gdb/value.c:1029
#4 0x555c4977be25 in allocate_value(type*) /home/simark/src/binutils-gdb/gdb/value.c:1040
#5 0x555c4979030d in value_primitive_field(value*, long, int, type*) /home/simark/src/binutils-gdb/gdb/value.c:3092
#6 0x555c478f6280 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:501
#7 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#8 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#9 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#10 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#11 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#12 0x555c4760d45f in c_value_print_struct /home/simark/src/binutils-gdb/gdb/c-valprint.c:383
#13 0x555c4760df4c in c_value_print_inner(value*, ui_file*, int, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:438
#14 0x555c483ff9a7 in language_defn::value_print_inner(value*, ui_file*, int, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:632
#15 0x555c49758b68 in do_val_print /home/simark/src/binutils-gdb/gdb/valprint.c:1048
#16 0x555c49759b17 in common_val_print(value*, ui_file*, int, value_print_options const*, language_defn const*) /home/simark/src/binutils-gdb/gdb/valprint.c:1151
#17 0x555c478f2fcb in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:335
#18 0x555c478f63b2 in cp_print_value /home/simark/src/binutils-gdb/gdb/cp-valprint.c:513
#19 0x555c478f02ca in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:161
#20 0x555c4760d45f in c_value_print_struct /home/simark/src/binutils-gdb/gdb/c-valprint.c:383
#21 0x555c4760df4c in c_value_print_inner(value*, ui_file*, int, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:438
#22 0x555c483ff9a7 in language_defn::value_print_inner(value*, ui_file*, int, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:632
#23 0x555c49758b68 in do_val_print /home/simark/src/binutils-gdb/gdb/valprint.c:1048
#24 0x555c49759b17 in common_val_print(value*, ui_file*, int, value_print_options const*, language_defn const*) /home/simark/src/binutils-gdb/gdb/valprint.c:1151
#25 0x555c478f2fcb in cp_print_value_fields(value*, ui_file*, int, value_print_options const*, type**, int) /home/simark/src/binutils-gdb/gdb/cp-valprint.c:335
#26 0x555c4760d45f in c_value_print_struct /home/simark/src/binutils-gdb/gdb/c-valprint.c:383
#27 0x555c4760df4c in c_value_print_inner(value*, ui_file*, int, value_print_options const*) /home/simark/src/binutils-gdb/gdb/c-valprint.c:438
#28 0x555c483ff9a7 in language_defn::value_print_inner(value*, ui_file*, int, value_print_options const*) const /home/simark/src/binutils-gdb/gdb/language.c:632
#29 0x555c49758b68 in do_val_print /home/simark/src/binutils-gdb/gdb/valprint.c:1048
Since there are some binaries with this in the wild, I think it would be
useful for GDB to work around this. I did the obvious simple thing, if
the DW_AT_data_member_location's value is -1, replace it with 0. I
added a producer check to only apply this fixup for GCC 11. The idea is
that if some other compiler ever uses a DW_AT_data_member_location value
of -1 by mistake, we don't know (before analyzing the bug at least) if
they did mean 0 or some other value. So I wouldn't want to apply the
fixup in that case.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=28063
Change-Id: Ieef3459b0b9bbce8bdad838ba83b4b64e7269d42
---
gdb/dwarf2/cu.c | 3 +-
gdb/dwarf2/cu.h | 1 +
gdb/dwarf2/read.c | 14 ++++
.../negative-data-member-location.c | 29 +++++++
.../negative-data-member-location.exp | 77 +++++++++++++++++++
5 files changed, 123 insertions(+), 1 deletion(-)
create mode 100644 gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
create mode 100644 gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
diff --git a/gdb/dwarf2/cu.c b/gdb/dwarf2/cu.c
index 019165f6e6c5..20aecd4da4d6 100644
--- a/gdb/dwarf2/cu.c
+++ b/gdb/dwarf2/cu.c
@@ -31,7 +31,8 @@ dwarf2_cu::dwarf2_cu (dwarf2_per_cu_data *per_cu,
has_loclist (false),
checked_producer (false),
producer_is_gxx_lt_4_6 (false),
- producer_is_gcc_lt_4_3 (false),
+ producer_is_gcc_lt_4_3 (false),
+ producer_is_gcc_11 (false),
producer_is_icc (false),
producer_is_icc_lt_14 (false),
producer_is_codewarrior (false),
diff --git a/gdb/dwarf2/cu.h b/gdb/dwarf2/cu.h
index 078479a56988..bce0a3de63a4 100644
--- a/gdb/dwarf2/cu.h
+++ b/gdb/dwarf2/cu.h
@@ -253,6 +253,7 @@ struct dwarf2_cu
bool checked_producer : 1;
bool producer_is_gxx_lt_4_6 : 1;
bool producer_is_gcc_lt_4_3 : 1;
+ bool producer_is_gcc_11 : 1;
bool producer_is_icc : 1;
bool producer_is_icc_lt_14 : 1;
bool producer_is_codewarrior : 1;
diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
index f7cb95b40cba..1a749eac334f 100644
--- a/gdb/dwarf2/read.c
+++ b/gdb/dwarf2/read.c
@@ -14320,6 +14320,7 @@ check_producer (struct dwarf2_cu *cu)
{
cu->producer_is_gxx_lt_4_6 = major < 4 || (major == 4 && minor < 6);
cu->producer_is_gcc_lt_4_3 = major < 4 || (major == 4 && minor < 3);
+ cu->producer_is_gcc_11 = major == 11;
}
else if (producer_is_icc (cu->producer, &major, &minor))
{
@@ -14464,6 +14465,19 @@ handle_member_location (struct die_info *die, struct dwarf2_cu *cu,
if (attr->form_is_constant ())
{
LONGEST offset = attr->constant_value (0);
+
+ /* Work around this GCC 11 bug, where it would erroneously use -1
+ data member locations, instead of 0:
+
+ Negative DW_AT_data_member_location
+ https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101378
+ */
+ if (offset == -1 && cu->producer_is_gcc_11)
+ {
+ complaint (_("DW_AT_data_member_location value of -1, assuming 0"));
+ offset = 0;
+ }
+
field->set_loc_bitpos (offset * bits_per_byte);
}
else if (attr->form_is_section_offset ())
diff --git a/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
new file mode 100644
index 000000000000..4871b2f12036
--- /dev/null
+++ b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
@@ -0,0 +1,29 @@
+/* Copyright (C) 2021 Free Software Foundation, Inc.
+
+ This file is part of GDB.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+struct the_struct
+{
+ char x[4];
+};
+
+struct the_struct s = { { 0x11, 0x22, 0x22, 0x11 } };
+
+int
+main (void)
+{
+ return 0;
+}
diff --git a/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
new file mode 100644
index 000000000000..9d05f208958b
--- /dev/null
+++ b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
@@ -0,0 +1,77 @@
+# Copyright 2021 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+# Test our workaround for a GCC 11 bug, where it sometimes puts a -1 value for
+# DW_AT_data_member_location:
+#
+# Negative DW_AT_data_member_location
+# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101378
+
+load_lib dwarf.exp
+
+# This test can only be run on targets which support DWARF-2 and use gas.
+if { ![dwarf2_support] } {
+ return 0
+}
+
+standard_testfile .c -dw.S
+
+set asm_file [standard_output_file ${srcfile2}]
+
+Dwarf::assemble ${asm_file} {
+ cu {} {
+ DW_TAG_compile_unit {
+ {DW_AT_language @DW_LANG_C99}
+ {DW_AT_producer "GNU C++17 11.1.0 -mtune=generic -march=x86-64 -g3 -O0"}
+ {name ${::srcfile}}
+ } {
+ declare_labels int_label struct_label
+
+ int_label: DW_TAG_base_type {
+ {DW_AT_byte_size 4 DW_FORM_udata}
+ {DW_AT_encoding @DW_ATE_signed}
+ {DW_AT_name "int"}
+ }
+
+ struct_label: DW_TAG_structure_type {
+ {DW_AT_name "the_struct"}
+ {DW_AT_byte_size 4 DW_FORM_udata}
+ } {
+ DW_TAG_member {
+ {DW_AT_name "field"}
+ {DW_AT_type :$int_label}
+ {DW_AT_data_member_location -1 DW_FORM_sdata}
+ }
+ }
+
+ DW_TAG_variable {
+ {DW_AT_name "s"}
+ {DW_AT_type :$struct_label}
+ {DW_AT_location {DW_OP_addr [gdb_target_symbol "s"]} SPECIAL_expr}
+ }
+ }
+ }
+}
+
+if { [prepare_for_testing "failed to prepare" ${testfile} [list $srcfile $asm_file] {nodebug}] } {
+ return
+}
+
+if { ![runto_main] } {
+ return
+}
+
+gdb_test "print /x s" " = {field = 0x11222211}"
+gdb_test "print /x s.field" " = 0x11222211"
base-commit: fdf1350dc13b66c0b2c9c961c28e37d993023d5a
--
2.34.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug
2022-01-26 19:55 [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug Simon Marchi
@ 2022-01-26 23:46 ` Lancelot SIX
2022-01-27 17:14 ` Simon Marchi
2022-01-27 20:45 ` Tom Tromey
1 sibling, 1 reply; 5+ messages in thread
From: Lancelot SIX @ 2022-01-26 23:46 UTC (permalink / raw)
To: Simon Marchi; +Cc: gdb-patches
Hi,
I have minor comments below:
> diff --git a/gdb/dwarf2/cu.c b/gdb/dwarf2/cu.c
> index 019165f6e6c5..20aecd4da4d6 100644
> --- a/gdb/dwarf2/cu.c
> +++ b/gdb/dwarf2/cu.c
> @@ -31,7 +31,8 @@ dwarf2_cu::dwarf2_cu (dwarf2_per_cu_data *per_cu,
> has_loclist (false),
> checked_producer (false),
> producer_is_gxx_lt_4_6 (false),
> - producer_is_gcc_lt_4_3 (false),
> + producer_is_gcc_lt_4_3 (false),
> + producer_is_gcc_11 (false),
^
You seem to be changing the indentation here (this changes 4 spaces for
1 tab).
> producer_is_icc (false),
> producer_is_icc_lt_14 (false),
> producer_is_codewarrior (false),
> diff --git a/gdb/dwarf2/cu.h b/gdb/dwarf2/cu.h
> index 078479a56988..bce0a3de63a4 100644
> --- a/gdb/dwarf2/cu.h
> +++ b/gdb/dwarf2/cu.h
> @@ -253,6 +253,7 @@ struct dwarf2_cu
> bool checked_producer : 1;
> bool producer_is_gxx_lt_4_6 : 1;
> bool producer_is_gcc_lt_4_3 : 1;
> + bool producer_is_gcc_11 : 1;
> bool producer_is_icc : 1;
> bool producer_is_icc_lt_14 : 1;
> bool producer_is_codewarrior : 1;
> diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
> index f7cb95b40cba..1a749eac334f 100644
> --- a/gdb/dwarf2/read.c
> +++ b/gdb/dwarf2/read.c
> @@ -14320,6 +14320,7 @@ check_producer (struct dwarf2_cu *cu)
> {
> cu->producer_is_gxx_lt_4_6 = major < 4 || (major == 4 && minor < 6);
> cu->producer_is_gcc_lt_4_3 = major < 4 || (major == 4 && minor < 3);
> + cu->producer_is_gcc_11 = major == 11;
> }
> else if (producer_is_icc (cu->producer, &major, &minor))
> {
> @@ -14464,6 +14465,19 @@ handle_member_location (struct die_info *die, struct dwarf2_cu *cu,
> if (attr->form_is_constant ())
> {
> LONGEST offset = attr->constant_value (0);
> +
> + /* Work around this GCC 11 bug, where it would erroneously use -1
> + data member locations, instead of 0:
> +
> + Negative DW_AT_data_member_location
> + https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101378
> + */
> + if (offset == -1 && cu->producer_is_gcc_11)
> + {
> + complaint (_("DW_AT_data_member_location value of -1, assuming 0"));
> + offset = 0;
> + }
> +
> field->set_loc_bitpos (offset * bits_per_byte);
> }
> else if (attr->form_is_section_offset ())
> diff --git a/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
> new file mode 100644
> index 000000000000..4871b2f12036
> --- /dev/null
> +++ b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
> @@ -0,0 +1,29 @@
> +/* Copyright (C) 2021 Free Software Foundation, Inc.
The year needs to be updated to 2021-2022.
> +
> + This file is part of GDB.
> +
> + This program is free software; you can redistribute it and/or modify
> + it under the terms of the GNU General Public License as published by
> + the Free Software Foundation; either version 3 of the License, or
> + (at your option) any later version.
> +
> + This program is distributed in the hope that it will be useful,
> + but WITHOUT ANY WARRANTY; without even the implied warranty of
> + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> + GNU General Public License for more details.
> +
> + You should have received a copy of the GNU General Public License
> + along with this program. If not, see <http://www.gnu.org/licenses/>. */
> +
> +struct the_struct
> +{
> + char x[4];
> +};
> +
> +struct the_struct s = { { 0x11, 0x22, 0x22, 0x11 } };
> +
> +int
> +main (void)
> +{
> + return 0;
> +}
> diff --git a/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
> new file mode 100644
> index 000000000000..9d05f208958b
> --- /dev/null
> +++ b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
> @@ -0,0 +1,77 @@
> +# Copyright 2021 Free Software Foundation, Inc.
Same, update the year from 2021 to 2021-2022.
Best,
Lancelot.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug
2022-01-26 23:46 ` Lancelot SIX
@ 2022-01-27 17:14 ` Simon Marchi
0 siblings, 0 replies; 5+ messages in thread
From: Simon Marchi @ 2022-01-27 17:14 UTC (permalink / raw)
To: Lancelot SIX; +Cc: gdb-patches
On 2022-01-26 18:46, Lancelot SIX wrote:
> Hi,
>
> I have minor comments below:
>
>> diff --git a/gdb/dwarf2/cu.c b/gdb/dwarf2/cu.c
>> index 019165f6e6c5..20aecd4da4d6 100644
>> --- a/gdb/dwarf2/cu.c
>> +++ b/gdb/dwarf2/cu.c
>> @@ -31,7 +31,8 @@ dwarf2_cu::dwarf2_cu (dwarf2_per_cu_data *per_cu,
>> has_loclist (false),
>> checked_producer (false),
>> producer_is_gxx_lt_4_6 (false),
>> - producer_is_gcc_lt_4_3 (false),
>> + producer_is_gcc_lt_4_3 (false),
>> + producer_is_gcc_11 (false),
> ^
>
> You seem to be changing the indentation here (this changes 4 spaces for
> 1 tab).
Ah, thanks. That's what happens when you work on multiple projects in
parallel, I had the wrong config loaded.
>> producer_is_icc (false),
>> producer_is_icc_lt_14 (false),
>> producer_is_codewarrior (false),
>> diff --git a/gdb/dwarf2/cu.h b/gdb/dwarf2/cu.h
>> index 078479a56988..bce0a3de63a4 100644
>> --- a/gdb/dwarf2/cu.h
>> +++ b/gdb/dwarf2/cu.h
>> @@ -253,6 +253,7 @@ struct dwarf2_cu
>> bool checked_producer : 1;
>> bool producer_is_gxx_lt_4_6 : 1;
>> bool producer_is_gcc_lt_4_3 : 1;
>> + bool producer_is_gcc_11 : 1;
>> bool producer_is_icc : 1;
>> bool producer_is_icc_lt_14 : 1;
>> bool producer_is_codewarrior : 1;
>> diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
>> index f7cb95b40cba..1a749eac334f 100644
>> --- a/gdb/dwarf2/read.c
>> +++ b/gdb/dwarf2/read.c
>> @@ -14320,6 +14320,7 @@ check_producer (struct dwarf2_cu *cu)
>> {
>> cu->producer_is_gxx_lt_4_6 = major < 4 || (major == 4 && minor < 6);
>> cu->producer_is_gcc_lt_4_3 = major < 4 || (major == 4 && minor < 3);
>> + cu->producer_is_gcc_11 = major == 11;
>> }
>> else if (producer_is_icc (cu->producer, &major, &minor))
>> {
>> @@ -14464,6 +14465,19 @@ handle_member_location (struct die_info *die, struct dwarf2_cu *cu,
>> if (attr->form_is_constant ())
>> {
>> LONGEST offset = attr->constant_value (0);
>> +
>> + /* Work around this GCC 11 bug, where it would erroneously use -1
>> + data member locations, instead of 0:
>> +
>> + Negative DW_AT_data_member_location
>> + https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101378
>> + */
>> + if (offset == -1 && cu->producer_is_gcc_11)
>> + {
>> + complaint (_("DW_AT_data_member_location value of -1, assuming 0"));
>> + offset = 0;
>> + }
>> +
>> field->set_loc_bitpos (offset * bits_per_byte);
>> }
>> else if (attr->form_is_section_offset ())
>> diff --git a/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
>> new file mode 100644
>> index 000000000000..4871b2f12036
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.c
>> @@ -0,0 +1,29 @@
>> +/* Copyright (C) 2021 Free Software Foundation, Inc.
>
> The year needs to be updated to 2021-2022.
Done.
>
>> +
>> + This file is part of GDB.
>> +
>> + This program is free software; you can redistribute it and/or modify
>> + it under the terms of the GNU General Public License as published by
>> + the Free Software Foundation; either version 3 of the License, or
>> + (at your option) any later version.
>> +
>> + This program is distributed in the hope that it will be useful,
>> + but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
>> + GNU General Public License for more details.
>> +
>> + You should have received a copy of the GNU General Public License
>> + along with this program. If not, see <http://www.gnu.org/licenses/>. */
>> +
>> +struct the_struct
>> +{
>> + char x[4];
>> +};
>> +
>> +struct the_struct s = { { 0x11, 0x22, 0x22, 0x11 } };
>> +
>> +int
>> +main (void)
>> +{
>> + return 0;
>> +}
>> diff --git a/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
>> new file mode 100644
>> index 000000000000..9d05f208958b
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.dwarf2/negative-data-member-location.exp
>> @@ -0,0 +1,77 @@
>> +# Copyright 2021 Free Software Foundation, Inc.
> Same, update the year from 2021 to 2021-2022.
Done.
Simon
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug
2022-01-26 19:55 [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug Simon Marchi
2022-01-26 23:46 ` Lancelot SIX
@ 2022-01-27 20:45 ` Tom Tromey
2022-01-27 20:46 ` Simon Marchi
1 sibling, 1 reply; 5+ messages in thread
From: Tom Tromey @ 2022-01-27 20:45 UTC (permalink / raw)
To: Simon Marchi via Gdb-patches
>>>>> "Simon" == Simon Marchi via Gdb-patches <gdb-patches@sourceware.org> writes:
Simon> has_loclist (false),
Simon> checked_producer (false),
Simon> producer_is_gxx_lt_4_6 (false),
Simon> - producer_is_gcc_lt_4_3 (false),
Simon> + producer_is_gcc_lt_4_3 (false),
Simon> + producer_is_gcc_11 (false),
Simon> producer_is_icc (false),
Simon> producer_is_icc_lt_14 (false),
The indentation looks weird here, but otherwise this seems good to me.
thanks,
Tom
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug
2022-01-27 20:45 ` Tom Tromey
@ 2022-01-27 20:46 ` Simon Marchi
0 siblings, 0 replies; 5+ messages in thread
From: Simon Marchi @ 2022-01-27 20:46 UTC (permalink / raw)
To: Tom Tromey, Simon Marchi via Gdb-patches
On 2022-01-27 15:45, Tom Tromey wrote:
>>>>>> "Simon" == Simon Marchi via Gdb-patches <gdb-patches@sourceware.org> writes:
>
> Simon> has_loclist (false),
> Simon> checked_producer (false),
> Simon> producer_is_gxx_lt_4_6 (false),
> Simon> - producer_is_gcc_lt_4_3 (false),
> Simon> + producer_is_gcc_lt_4_3 (false),
> Simon> + producer_is_gcc_11 (false),
> Simon> producer_is_icc (false),
> Simon> producer_is_icc_lt_14 (false),
>
> The indentation looks weird here, but otherwise this seems good to me.
Yes, Lancelot pointed it out.
I sent this as a result :)
https://sourceware.org/pipermail/gdb-patches/2022-January/185511.html
Simon
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2022-01-27 20:46 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-01-26 19:55 [PATCH v2] gdb: work around negative DW_AT_data_member_location GCC 11 bug Simon Marchi
2022-01-26 23:46 ` Lancelot SIX
2022-01-27 17:14 ` Simon Marchi
2022-01-27 20:45 ` Tom Tromey
2022-01-27 20:46 ` Simon Marchi
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).