[v2,2/2,gdb/exp] Fix ignoring of incorrect namespace prefix
Checks
| Context |
Check |
Description |
| linaro-tcwg-bot/tcwg_gdb_build--master-aarch64 |
success
|
Build passed
|
| linaro-tcwg-bot/tcwg_gdb_build--master-arm |
success
|
Build passed
|
| linaro-tcwg-bot/tcwg_gdb_check--master-arm |
success
|
Test passed
|
| linaro-tcwg-bot/tcwg_gdb_check--master-aarch64 |
success
|
Test passed
|
Commit Message
Consider test.c, compiled to a.out using "g++ -g test.c":
...
1 namespace mod_a { int xxx = 10; }
2 namespace mod_b { using namespace mod_a;
3 int yyy = 20; }
4 int main (void) {
5 using namespace mod_b;
6 void (xxx + yyy);
7 return 0;
8 }
...
When trying to print the value of non-existent variable mod_a::yyy, we get:
...
$ gdb -q -batch a.out -ex start -ex "print mod_a::yyy"
...
Temporary breakpoint 1, main () at test.c:7
7 return 0;
$1 = 20
...
The problem is in cp_lookup_symbol_via_imports, where we decide that the
"using namespace mod_b" from main is applicable in scope mod_a.
Fix this by being more strict in the calculation of directive_match:
...
if (len == 0)
- directive_match = 1;
+ {
+ const char *current_scope = (block->function () != nullptr
+ ? block->scope ()
+ : nullptr /* Don't know. */);
+ directive_match = (current_scope != nullptr
+ ? streq (scope, current_scope)
+ : 1 /* Assume there's a match. */);
+ }
...
As is clear from the code, in case we don't know the current scope, we assume
there's a match. This may be harmless, or this may describe a cornercase we
haven't run into yet. If so, it's a pre-existing issue.
The new test-case contains regression tests for:
- PR34051, and
- PR34034 for which it contains a kfail.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=34051
---
gdb/cp-namespace.c | 9 +++-
gdb/testsuite/gdb.cp/nsusing-2.cc | 39 ++++++++++++++++
gdb/testsuite/gdb.cp/nsusing-2.exp | 75 ++++++++++++++++++++++++++++++
3 files changed, 122 insertions(+), 1 deletion(-)
create mode 100644 gdb/testsuite/gdb.cp/nsusing-2.cc
create mode 100644 gdb/testsuite/gdb.cp/nsusing-2.exp
Comments
Hello Tom,
No comment on the merit of the patch, just a couple of nits regarding
the testcase:
Tom de Vries <tdevries@suse.de> writes:
> diff --git a/gdb/testsuite/gdb.cp/nsusing-2.exp b/gdb/testsuite/gdb.cp/nsusing-2.exp
> new file mode 100644
> index 00000000000..9653616f7f4
> --- /dev/null
> +++ b/gdb/testsuite/gdb.cp/nsusing-2.exp
> @@ -0,0 +1,75 @@
> +# Copyright 2026 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 recursive "using namespace". Regression test for PR34034 and PR34051.
> +
> +standard_testfile .cc
> +
> +if {[prepare_for_testing "failed to prepare" $testfile $srcfile \
> + {debug c++}]} {
> + return -1
AFAIU there's no need to return a value from the toplevel of a testcase,
and current style is to have just a "return" here.
> +}
> +
> +with_test_prefix pre-main {
> + gdb_test "print mod_a::xxx" " = 10"
> + gdb_test "print mod_b::yyy" " = 20"
> +
> + # Namespace mod_b is using namespace mod_a, so mod_a::xxx is available as
> + # mod_b::xxx. This is not available here though, but later, at
> + # start-of-main. I wonder if this should also be availabe here.
Typo: available
> + gdb_test "print mod_b::xxx" \
> + [string_to_regexp {No symbol "xxx" in namespace "mod_b".}]
> +}
On 4/9/26 8:30 AM, Thiago Jung Bauermann wrote:
> Hello Tom,
>
> No comment on the merit of the patch, just a couple of nits regarding
> the testcase:
Hi Thiago,
thanks for the review.
I've fixed the two issues you mentioned, and submitted a v3 (
https://sourceware.org/pipermail/gdb-patches/2026-April/226433.html ).
Thanks,
- Tom
> Tom de Vries <tdevries@suse.de> writes:
>
>> diff --git a/gdb/testsuite/gdb.cp/nsusing-2.exp b/gdb/testsuite/gdb.cp/nsusing-2.exp
>> new file mode 100644
>> index 00000000000..9653616f7f4
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.cp/nsusing-2.exp
>> @@ -0,0 +1,75 @@
>> +# Copyright 2026 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 recursive "using namespace". Regression test for PR34034 and PR34051.
>> +
>> +standard_testfile .cc
>> +
>> +if {[prepare_for_testing "failed to prepare" $testfile $srcfile \
>> + {debug c++}]} {
>> + return -1
>
> AFAIU there's no need to return a value from the toplevel of a testcase,
> and current style is to have just a "return" here.
>
>> +}
>> +
>> +with_test_prefix pre-main {
>> + gdb_test "print mod_a::xxx" " = 10"
>> + gdb_test "print mod_b::yyy" " = 20"
>> +
>> + # Namespace mod_b is using namespace mod_a, so mod_a::xxx is available as
>> + # mod_b::xxx. This is not available here though, but later, at
>> + # start-of-main. I wonder if this should also be availabe here.
>
> Typo: available
>
>> + gdb_test "print mod_b::xxx" \
>> + [string_to_regexp {No symbol "xxx" in namespace "mod_b".}]
>> +}
>
@@ -431,7 +431,14 @@ cp_lookup_symbol_via_imports (const char *scope,
if (search_parents)
{
if (len == 0)
- directive_match = 1;
+ {
+ const char *current_scope = (block->function () != nullptr
+ ? block->scope ()
+ : nullptr /* Don't know. */);
+ directive_match = (current_scope != nullptr
+ ? streq (scope, current_scope)
+ : 1 /* Assume there's a match. */);
+ }
else
directive_match = (startswith (scope, current->import_dest)
&& (scope[len] == ':'
new file mode 100644
@@ -0,0 +1,39 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+ Copyright 2026 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/>. */
+
+/* C++ variant of the Fortran example from PR34034. */
+
+namespace mod_a {
+ int xxx = 10;
+}
+
+namespace mod_b {
+ using namespace mod_a;
+ int yyy = 20;
+}
+
+static void foo () {}
+
+int
+main (void)
+{
+ foo (); /* main-entry. */
+ using namespace mod_b;
+ (void)xxx;
+ (void)yyy;
+ return 0; /* main-return. */
+}
new file mode 100644
@@ -0,0 +1,75 @@
+# Copyright 2026 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 recursive "using namespace". Regression test for PR34034 and PR34051.
+
+standard_testfile .cc
+
+if {[prepare_for_testing "failed to prepare" $testfile $srcfile \
+ {debug c++}]} {
+ return -1
+}
+
+with_test_prefix pre-main {
+ gdb_test "print mod_a::xxx" " = 10"
+ gdb_test "print mod_b::yyy" " = 20"
+
+ # Namespace mod_b is using namespace mod_a, so mod_a::xxx is available as
+ # mod_b::xxx. This is not available here though, but later, at
+ # start-of-main. I wonder if this should also be availabe here.
+ gdb_test "print mod_b::xxx" \
+ [string_to_regexp {No symbol "xxx" in namespace "mod_b".}]
+}
+
+set line_main_entry [gdb_get_line_number main-entry]
+if {![runto $srcfile:$line_main_entry]} {
+ return
+}
+
+# Start of main. Function main is not yet using namespace mod_b.
+with_test_prefix start-of-main {
+ # Namespace mod_b is using namespace mod_a, so mod_a::xxx is available as
+ # mod_b::xxx. See also the note at the identical command in pre-main.
+ gdb_test "print mod_b::xxx" " = 10"
+
+ # Same command as in end-of-main, but not a regression test for PR34034.
+ gdb_test "print xxx" \
+ [string_to_regexp {No symbol "xxx" in current context.}]
+
+ # Same test as in end-of-main, but not a regression test for PR34051.
+ gdb_test "print mod_a::yyy" \
+ [string_to_regexp {No symbol "yyy" in namespace "mod_a".}]
+}
+
+set line_main_return [gdb_get_line_number "main-return"]
+gdb_test "next" \
+ [subst_vars {$line_main_return\t[^\r\n]+}]
+
+# End of main. Function main is using namespace mod_b.
+with_test_prefix end-of-main {
+ # Function main is using namespace mod_b, so mod_b::yyy is available as
+ # yyy.
+ gdb_test "print yyy" " = 20"
+
+ # Function main is using namespace mod_b, and namespace mod_b is using
+ # namespace mod_a, so mod_a::xxx is available as xxx. Regression test for
+ # PR34034.
+ setup_kfail exp/34034 *-*-*
+ gdb_test "print xxx" " = 10"
+
+ # This used to print " $<n> = 20". Regression test for PR34051.
+ gdb_test "print mod_a::yyy" \
+ [string_to_regexp {No symbol "yyy" in namespace "mod_a".}]
+}