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] [gdb/exp] Limit workaround in using_direct::valid_line to broken GCC versions
Date: Thu,  4 Jun 2026 18:21:35 +0200	[thread overview]
Message-ID: <20260604162135.2846175-1-tdevries@suse.de> (raw)

Consider test.c:
...
     1	namespace mod_a {
     2	  int xxx = 10;
     3	}
     4
     5	static void
     6	foo ()
     7	{
     8	}
     9
    10	int
    11	main ()
    12	{
    13	  {
    14	    foo ();
    15	    using namespace mod_a;
    16	  }
    17
    18	  return mod_a::xxx;
    19	}
...
compiled with "g++ test.c -g".

Attempting to print xxx at line 14 shouldn't find anything (because it's
before the "using namespace mod_a"), but it does:
...
$ gdb -q -batch a.out -ex start -ex "p xxx"
...
Temporary breakpoint 1, main () at test.c:14
14	    foo ();
$1 = 10
...

This happens because using_direct::valid_line returns true here:
...
      return (decl_line <= curr_sal.line)
	     || (decl_line >= boundary);
...

Since we have decl_line == 15 and curr_sal.line == 14,
"(decl_line <= curr_sal.line)" evaluates to false.

But boundary == 14, so "(decl_line >= boundary)" evaluates to true.

The "(decl_line >= boundary)" bit was added as a workaround for
GCC PR debug/108716.

Since I'm using GCC 15, the workaround is not needed.

Fix this by limiting the workaround to broken GCC versions.

Tested on x86_64-linux, using GCC 15.2, and 7.5.

Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=34203
---
 gdb/cp-namespace.c                       | 26 +++++++++++---
 gdb/namespace.c                          |  9 +++--
 gdb/testsuite/gdb.cp/ns-before-using.c   | 37 ++++++++++++++++++++
 gdb/testsuite/gdb.cp/ns-before-using.exp | 44 ++++++++++++++++++++++++
 4 files changed, 110 insertions(+), 6 deletions(-)
 create mode 100644 gdb/testsuite/gdb.cp/ns-before-using.c
 create mode 100644 gdb/testsuite/gdb.cp/ns-before-using.exp

diff --git a/gdb/cp-namespace.c b/gdb/cp-namespace.c
index c8cd5c245aa..16fd4e2ded8 100644
--- a/gdb/cp-namespace.c
+++ b/gdb/cp-namespace.c
@@ -33,6 +33,7 @@
 #include "namespace.h"
 #include "inferior.h"
 #include "gdbsupport/unordered_map.h"
+#include "producer.h"
 #include <string>
 #include <string.h>
 
@@ -410,9 +411,26 @@ cp_lookup_symbol_via_imports (const char *scope,
 	found_symbols[sym.symbol->m_name] = sym;
     }
 
-  /* Due to a GCC bug, we need to know the boundaries of the current block
-     to know if a certain using directive is valid.  */
-  symtab_and_line boundary_sal = find_sal_for_pc (block->end () - 1, 0);
+  unsigned boundary_line = 0;
+  {
+    struct symbol *fn = block->containing_function ();
+    int major, minor;
+    if (fn != nullptr
+	&& producer_is_gcc (fn->symtab ()->compunit ()->producer (),
+			    &major, &minor)
+	&& (major <= 9
+	    || (major == 10 && minor < 5)
+	    || (major == 11 && minor < 4)
+	    || (major == 12 && minor < 3)
+	    || (major == 13 && minor < 1)))
+      {
+	/* Due to a GCC bug (PR debug/108716, fixed in 10.5, 11.4, 12.3, 13.1),
+	   we need to know the boundaries of the current block to know if a
+	   certain using directive is valid.  */
+	symtab_and_line boundary_sal = find_sal_for_pc (block->end () - 1, 0);
+	boundary_line = boundary_sal.line;
+      }
+  }
 
   /* Go through the using directives.  If any of them add new names to
      the namespace we're searching in, see if we can find a match by
@@ -423,7 +441,7 @@ cp_lookup_symbol_via_imports (const char *scope,
 
       /* If the using directive was below the place we are stopped at,
 	 do not use this directive.  */
-      if (!current->valid_line (boundary_sal.line))
+      if (!current->valid_line (boundary_line))
 	continue;
       len = strlen (current->import_dest);
       directive_match = (search_parents
diff --git a/gdb/namespace.c b/gdb/namespace.c
index 5b5749369ee..d81365a284a 100644
--- a/gdb/namespace.c
+++ b/gdb/namespace.c
@@ -113,8 +113,13 @@ using_direct::valid_line (unsigned int boundary) const
     {
       CORE_ADDR curr_pc = get_frame_pc (get_selected_frame (nullptr));
       symtab_and_line curr_sal = find_sal_for_pc (curr_pc, 0);
-      return (decl_line <= curr_sal.line)
-	     || (decl_line >= boundary);
+
+      /* Apply GCC PR debug/108716 workaround.  */
+      if (boundary != 0
+	  && decl_line >= boundary)
+	return true;
+
+      return decl_line <= curr_sal.line;
     }
   catch (const gdb_exception &ex)
     {
diff --git a/gdb/testsuite/gdb.cp/ns-before-using.c b/gdb/testsuite/gdb.cp/ns-before-using.c
new file mode 100644
index 00000000000..f74c35b0d3f
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/ns-before-using.c
@@ -0,0 +1,37 @@
+/* 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/>.  */
+
+namespace mod_a
+{
+  int xxx = 10;
+}
+
+static void
+foo ()
+{
+}
+
+int
+main ()
+{
+  {
+    foo ();
+    using namespace mod_a;
+  }
+
+  return mod_a::xxx;
+}
diff --git a/gdb/testsuite/gdb.cp/ns-before-using.exp b/gdb/testsuite/gdb.cp/ns-before-using.exp
new file mode 100644
index 00000000000..f891cf56c49
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/ns-before-using.exp
@@ -0,0 +1,44 @@
+# 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/>.
+
+standard_testfile .c
+
+if {[prepare_for_testing "failed to prepare" $testfile $srcfile \
+	 {debug c++}]} {
+    return
+}
+
+if {![runto_main]} {
+    continue
+}
+
+# Xfail for incorrect decl_line on DW_TAG_imported_module,
+# GCC PR debug/108716.
+set have_gcc108716_xfail \
+    [expr {[test_compiler_info gcc-*] && [gcc_major_version] < 13}]
+
+# Regression test for PR exp/34203.
+set re_pass [string_to_regexp {No symbol "xxx" in current context.}]
+gdb_test_multiple "print xxx" "" {
+    -re -wrap $re_pass {
+	pass $gdb_test_name
+    }
+    -re -wrap "$valnum_re = 10" {
+	if {$have_gcc108716_xfail} {
+	    setup_xfail *-*-* gcc/108716
+	}
+	fail $gdb_test_name
+    }
+}

base-commit: bd64797371d27c766d551d0bf115d9090f1d0594
-- 
2.51.0


             reply	other threads:[~2026-06-04 16:22 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-04 16:21 Tom de Vries [this message]
2026-06-18  7:53 ` [PING][PATCH] " Tom de Vries
2026-07-13 16:17   ` 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=20260604162135.2846175-1-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