[pushed] PR breakpoints/17000: user breakpoint not inserted if software-single-step at same location - another test

[Pedro Alves <palves at redhat dot com>]

On 06/03/2014 09:22 AM, Pedro Alves wrote:
> On 06/03/2014 12:16 AM, Pedro Alves wrote:
> 
>> Let me know what you think, and feel free to push if it
>> looks OK.  Tested on all combinations of
>> native|gdbserver X hardware-|software- stepping.  At least,
>> I think I did.  If not this exact version, some other minor
>> variation.  :-)
> 
> Bah, I woke up realizing that the version I posted forgets to
> clone the shadow buffer!  Let me fix that and repost...
> 

I managed to come up with a test that exposed this.  Basically, it does:

(gdb) disassemble test
Dump of assembler code for function test:
   0x0000000000400766 <+0>:     push   %rbp
   0x0000000000400767 <+1>:     mov    %rsp,%rbp
=> 0x000000000040076a <+4>:     nop
   0x000000000040076b <+5>:     nop
   0x000000000040076c <+6>:     pop    %rbp
   0x000000000040076d <+7>:     retq   
End of assembler dump.

(gdb) si&
(gdb) del $bpnum

(gdb) disassemble test
Dump of assembler code for function test:
   0x0000000000400766 <+0>:     push   %rbp
   0x0000000000400767 <+1>:     mov    %rsp,%rbp
   0x000000000040076a <+4>:     nop
=> 0x000000000040076b <+5>:     int3   
   0x000000000040076c <+6>:     pop    %rbp
   0x000000000040076d <+7>:     retq   
End of assembler dump.

And note the bogus "int3" in the second disassemble output.

The test actually fails in a different way currently against
gdbserver:

(gdb) PASS: gdb.base/sss-bp-on-user-bp-2.exp: define stepi_del_break
stepi_del_break
Cannot execute this command while the target is running.
(gdb) FAIL: gdb.base/sss-bp-on-user-bp-2.exp: stepi_del_break

The error is because GDB tried to remove the breakpoint from
memory while the thread was running, and we can't talk to the
server in the all-stop RSP until we get a stop reply, but in any
case, GDB shouldn't even be attempting to remove the breakpoint,
exactly because there was a sss breakpoint at the same address.

I've added a kfail, and pushed it.

8<--------------
PR breakpoints/17000: user breakpoint not inserted if software-single-step at same location - test

GDB gets confused when removing a software single-step breakpoint that
is at the same address as another breakpoint.  Add another kfailed
test.

gdb/testsuite/
2014-06-03  Pedro Alves  <palves@redhat.com>

	PR breakpoints/17000
	* gdb.base/sss-bp-on-user-bp-2.c: New file.
	* gdb.base/sss-bp-on-user-bp-2.exp: New file.
---
 gdb/testsuite/ChangeLog                        |   6 ++
 gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.c   |  29 +++++++
 gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.exp | 109 +++++++++++++++++++++++++
 3 files changed, 144 insertions(+)
 create mode 100644 gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.c
 create mode 100644 gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.exp

diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index 9ded6eb..644f5ac 100644
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,3 +1,9 @@
+2014-06-03  Pedro Alves  <palves@redhat.com>
+
+	PR breakpoints/17000
+	* gdb.base/sss-bp-on-user-bp-2.c: New file.
+	* gdb.base/sss-bp-on-user-bp-2.exp: New file.
+
 2014-06-02  Doug Evans  <xdje42@gmail.com>
 
 	* gdb.guile/scm-parameter.exp: New file.
diff --git a/gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.c b/gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.c
new file mode 100644
index 0000000..c3c26fe
--- /dev/null
+++ b/gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.c
@@ -0,0 +1,29 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2014 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/>.  */
+
+static void
+test (void)
+{
+  label: asm ("  nop"); label2: asm ("  nop"); /* must be a single line */
+}
+
+int
+main (void)
+{
+  test ();
+  return 0;
+}
diff --git a/gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.exp b/gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.exp
new file mode 100644
index 0000000..a129bb7
--- /dev/null
+++ b/gdb/testsuite/gdb.base/sss-bp-on-user-bp-2.exp
@@ -0,0 +1,109 @@
+# Copyright (C) 2014 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 that GDB doesn't get confused in the following scenario
+# (PR breakpoints/17000).  Say, we have this program:
+#
+#  =>  0xff000001  INSN1
+#      0xff000002  INSN2
+#
+# The PC currently points at INSN1.
+#
+# 1 - User sets a breakpoint at 0xff000002 (INSN2).
+#
+# 2 - User steps.  On software single-step archs, this sets a software
+#     single-step breakpoint at 0xff000002 (INSN2) too.
+#
+# 3 - User deletes breakpoint (INSN2) before the single-step finishes.
+#
+# 4 - The single-step finishes, and GDB removes the single-step
+#     breakpoint.
+
+standard_testfile
+
+if {[prepare_for_testing "failed to prepare" $testfile $srcfile debug]} {
+    return -1
+}
+
+if ![runto_main] {
+    fail "Can't run to main"
+    return 0
+}
+
+set line_re "\[^\r\n\]*"
+
+gdb_test "b test:label" "Breakpoint .*"
+gdb_continue_to_breakpoint "run past setup"
+delete_breakpoints
+
+# So we can precisely control breakpoint insertion order.
+gdb_test_no_output "set breakpoint always-inserted on"
+
+# Capture disassembly output.  PREFIX is used as test prefix.  The
+# current instruction indicator (=>) is stripped away.
+proc disassemble { prefix } {
+    with_test_prefix "$prefix" {
+	set output [capture_command_output "disassemble test" ""]
+	return [string map {"=>" "  "} $output]
+    }
+}
+
+# Issue a stepi and immediately delete the user breakpoint that is set
+# at the same address as the software single-step breakpoint.  Do this
+# in a user defined command, so that the stepi's trap doesn't have a
+# chance to be handled before further input is processed.  We then
+# compare before/after disassembly.  GDB should be able to handle
+# deleting the user breakpoint before deleting the single-step
+# breakpoint.  E.g., we shouldn't see breakpoint instructions in the
+# disassembly.
+
+set disasm_before [disassemble "before"]
+
+gdb_test "b test:label2" ".*" "set breakpoint where si will land"
+
+set test "define stepi_del_break"
+gdb_test_multiple $test $test {
+    -re "Type commands for definition of \"stepi_del_break\".\r\nEnd with a line saying just \"end\".\r\n>$" {
+	gdb_test "si&\ndel \$bpnum\nend" "" $test
+    }
+}
+
+set command "stepi_del_break"
+set test $command
+setup_kfail "breakpoints/17000" "*-*-*"
+gdb_test_multiple $command $test {
+    -re "^$command\r\n$gdb_prompt " {
+	# Note no end anchor, because "si&" finishes and prints the
+	# current frame/line after the prompt is printed.
+	pass $test
+    }
+}
+
+# Now consume the output of the finished "si&".
+set test "si& finished"
+gdb_test_multiple "" $test {
+    -re "must be a single line \\\*/\r\n" {
+	pass $test
+    }
+}
+
+set disasm_after [disassemble "after"]
+
+set test "before/after disassembly matches"
+if ![string compare $disasm_before $disasm_after]  {
+    pass $test
+} else {
+    fail $test
+}
-- 
1.9.0