[PATCH] Match demangled name in "skip"

[PATCH] Match demangled name in "skip"

[Tom Tromey]

PR gdb/26598 notes that, before commit bcfe6157ca28 ("Use the linkage
name if it exists"), the "skip" command would match the demangled name
of a symbol, but now only matches the linkage name.

This patch fixes this regression.  I looked at all calls to
function_name_is_marked_for_skip, and only one used the linkage name.

gdb/ChangeLog
2020-09-16  Tom Tromey  <tromey@adacore.com>

	PR gdb/26598:
	* infrun.c (fill_in_stop_func): Use find_pc_partial_function_sym.

gdb/testsuite/ChangeLog
2020-09-16  Tom Tromey  <tromey@adacore.com>

	PR gdb/26598:
	* gdb.base/skipcxx.exp: New file.
	* gdb.base/skipcxx.cc: New file.
---
 gdb/ChangeLog                      |  5 +++++
 gdb/infrun.c                       | 12 ++++++-----
 gdb/testsuite/ChangeLog            |  6 ++++++
 gdb/testsuite/gdb.base/skipcxx.cc  | 32 ++++++++++++++++++++++++++++++
 gdb/testsuite/gdb.base/skipcxx.exp | 29 +++++++++++++++++++++++++++
 5 files changed, 79 insertions(+), 5 deletions(-)
 create mode 100644 gdb/testsuite/gdb.base/skipcxx.cc
 create mode 100644 gdb/testsuite/gdb.base/skipcxx.exp

diff --git a/gdb/infrun.c b/gdb/infrun.c
index 5773fd03951..780d5bc791e 100644
--- a/gdb/infrun.c
+++ b/gdb/infrun.c
@@ -4358,14 +4358,16 @@ fill_in_stop_func (struct gdbarch *gdbarch,
   if (!ecs->stop_func_filled_in)
     {
       const block *block;
+      const general_symbol_info *gsi;
 
       /* Don't care about return value; stop_func_start and stop_func_name
 	 will both be 0 if it doesn't work.  */
-      find_pc_partial_function (ecs->event_thread->suspend.stop_pc,
-				&ecs->stop_func_name,
-				&ecs->stop_func_start,
-				&ecs->stop_func_end,
-				&block);
+      find_pc_partial_function_sym (ecs->event_thread->suspend.stop_pc,
+				    &gsi,
+				    &ecs->stop_func_start,
+				    &ecs->stop_func_end,
+				    &block);
+      ecs->stop_func_name = gsi == nullptr ? nullptr : gsi->print_name ();
 
       /* The call to find_pc_partial_function, above, will set
 	 stop_func_start and stop_func_end to the start and end
diff --git a/gdb/testsuite/gdb.base/skipcxx.cc b/gdb/testsuite/gdb.base/skipcxx.cc
new file mode 100644
index 00000000000..5f653455c36
--- /dev/null
+++ b/gdb/testsuite/gdb.base/skipcxx.cc
@@ -0,0 +1,32 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2020 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 somename
+{
+int func()
+{
+  return 23;
+}
+}
+
+int
+main ()
+{
+  int x = somename::func ();
+
+  return 0;
+}
diff --git a/gdb/testsuite/gdb.base/skipcxx.exp b/gdb/testsuite/gdb.base/skipcxx.exp
new file mode 100644
index 00000000000..19144519eee
--- /dev/null
+++ b/gdb/testsuite/gdb.base/skipcxx.exp
@@ -0,0 +1,29 @@
+#   Copyright 2020 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 .cc
+
+if {[prepare_for_testing "failed to prepare" $testfile $srcfile debug]} {
+    return -1
+}
+
+if ![runto_main] {
+    fail "can't run to main"
+    return
+}
+
+gdb_test "skip -rfu ^somename::" \
+    [string_to_regexp "Function(s) ^somename:: will be skipped when stepping."]
+gdb_test "step" ".* return 0;"
-- 
2.26.2

Re: [PATCH] Match demangled name in "skip"

[Simon Marchi]

On 2020-09-16 2:10 p.m., Tom Tromey wrote:
> PR gdb/26598 notes that, before commit bcfe6157ca28 ("Use the linkage
> name if it exists"), the "skip" command would match the demangled name
> of a symbol, but now only matches the linkage name.
>
> This patch fixes this regression.  I looked at all calls to
> function_name_is_marked_for_skip, and only one used the linkage name.
>
> gdb/ChangeLog
> 2020-09-16  Tom Tromey  <tromey@adacore.com>
>
> 	PR gdb/26598:
> 	* infrun.c (fill_in_stop_func): Use find_pc_partial_function_sym.
>
> gdb/testsuite/ChangeLog
> 2020-09-16  Tom Tromey  <tromey@adacore.com>
>
> 	PR gdb/26598:
> 	* gdb.base/skipcxx.exp: New file.
> 	* gdb.base/skipcxx.cc: New file.

Thanks, that LGTM.

Since this changes stop_func_name from the mangled name to the demangled
name (like it was before your change?), I checked what else
stop_func_name was used for.  It is passed to
gdbarch_in_solib_return_trampoline.  Implementations use that name to
compare against hard-coded symbol names.  However, I would guess that
these symbol names are not prone to being mangled.  Otherwise, we'd need
to save both the demangled and linkage name.

Simon

Re: [PATCH] Match demangled name in "skip"

[Tom Tromey]

>>>>> "Simon" == Simon Marchi <simark@simark.ca> writes:

>> gdb/testsuite/ChangeLog
>> 2020-09-16  Tom Tromey  <tromey@adacore.com>
>> 
>> PR gdb/26598:
>> * gdb.base/skipcxx.exp: New file.
>> * gdb.base/skipcxx.cc: New file.

Simon> Thanks, that LGTM.

Thanks, I'm going to check it in, and merge it to gdb-10 as well.

Simon> Since this changes stop_func_name from the mangled name to the demangled
Simon> name (like it was before your change?), I checked what else
Simon> stop_func_name was used for.

Thanks for doing this, I missed that.

Tom