Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Tom de Vries <tdevries@suse.de>
To: gdb-patches@sourceware.org
Subject: [PATCH v3 2/2] [gdb/exp] Fix ignoring of incorrect namespace prefix
Date: Fri, 10 Apr 2026 14:11:44 +0200	[thread overview]
Message-ID: <20260410121144.1836026-3-tdevries@suse.de> (raw)
In-Reply-To: <20260410121144.1836026-1-tdevries@suse.de>

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

diff --git a/gdb/cp-namespace.c b/gdb/cp-namespace.c
index f8b71be3ff1..45cde597674 100644
--- a/gdb/cp-namespace.c
+++ b/gdb/cp-namespace.c
@@ -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] == ':'
diff --git a/gdb/testsuite/gdb.cp/nsusing-2.cc b/gdb/testsuite/gdb.cp/nsusing-2.cc
new file mode 100644
index 00000000000..72f98add1b9
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/nsusing-2.cc
@@ -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.  */
+}
diff --git a/gdb/testsuite/gdb.cp/nsusing-2.exp b/gdb/testsuite/gdb.cp/nsusing-2.exp
new file mode 100644
index 00000000000..647a8600955
--- /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
+}
+
+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 available 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".}]
+}
-- 
2.51.0


  parent reply	other threads:[~2026-04-10 12:13 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-10 12:11 [PATCH v3 0/2] " Tom de Vries
2026-04-10 12:11 ` [PATCH v3 1/2] [gdb] Break up complex assignment in cp_lookup_symbol_via_imports Tom de Vries
2026-04-14 17:49   ` Tom Tromey
2026-04-10 12:11 ` Tom de Vries [this message]
2026-04-14 18:26   ` [PATCH v3 2/2] [gdb/exp] Fix ignoring of incorrect namespace prefix Tom Tromey
2026-04-15 10:12     ` Tom de Vries

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260410121144.1836026-3-tdevries@suse.de \
    --to=tdevries@suse.de \
    --cc=gdb-patches@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox