Re: [PATCH v2] Add mi-threads-interrupt.exp test (PR 20039)

Re: [PATCH v2] Add mi-threads-interrupt.exp test (PR 20039)

[Pedro Alves]

On 05/05/2016 07:19 PM, Simon Marchi wrote:
> Add a new test for PR 20039.  The test spawns new threads, then tries to
> interrupt, continue, and interrupt again.  This use case was fixed by
> commit 5fe966540d6b748f825774868463003700f0c878 in master, but gdb 7.11
> is affected (so if you try it on the gdb-7.11-branch right now, the test
> will fail).
> 
> New in v2, the test now handles mi-async on mode properly.  The failure
> was specific to mi-async off, but I don't think it's bad to test the
> same thing under async on mode.  I added a little hack when running in
> async mode to work around bug 20045.
> 
> I also removed one continue/interrupt pair, as a single one was enough to
> trigger the problem.
> 
> gdb/testsuite/ChangeLog:
> 
> 	* gdb.mi/mi-threads-interrupt.c: New file.
> 	* gdb.mi/mi-threads-interrupt.exp: New file.

OK.

Thanks,
Pedro Alves

Re: [PATCH v2] Add mi-threads-interrupt.exp test (PR 20039)

[Simon Marchi]

On 16-05-05 02:19 PM, Simon Marchi wrote:
> Add a new test for PR 20039.  The test spawns new threads, then tries to
> interrupt, continue, and interrupt again.  This use case was fixed by
> commit 5fe966540d6b748f825774868463003700f0c878 in master, but gdb 7.11
> is affected (so if you try it on the gdb-7.11-branch right now, the test
> will fail).
> 
> New in v2, the test now handles mi-async on mode properly.  The failure
> was specific to mi-async off, but I don't think it's bad to test the
> same thing under async on mode.  I added a little hack when running in
> async mode to work around bug 20045.
> 
> I also removed one continue/interrupt pair, as a single one was enough to
> trigger the problem.
> 
> gdb/testsuite/ChangeLog:
> 
> 	* gdb.mi/mi-threads-interrupt.c: New file.
> 	* gdb.mi/mi-threads-interrupt.exp: New file.
> ---
>  gdb/testsuite/gdb.mi/mi-threads-interrupt.c   | 61 ++++++++++++++++++++++
>  gdb/testsuite/gdb.mi/mi-threads-interrupt.exp | 75 +++++++++++++++++++++++++++
>  2 files changed, 136 insertions(+)
>  create mode 100644 gdb/testsuite/gdb.mi/mi-threads-interrupt.c
>  create mode 100644 gdb/testsuite/gdb.mi/mi-threads-interrupt.exp
> 
> diff --git a/gdb/testsuite/gdb.mi/mi-threads-interrupt.c b/gdb/testsuite/gdb.mi/mi-threads-interrupt.c
> new file mode 100644
> index 0000000..8747806
> --- /dev/null
> +++ b/gdb/testsuite/gdb.mi/mi-threads-interrupt.c
> @@ -0,0 +1,61 @@
> +/* This testcase is part of GDB, the GNU debugger.
> +
> +   Copyright 2016 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/>.  */
> +
> +#include <pthread.h>
> +#include <unistd.h>
> +
> +#define NUM_THREADS 2
> +
> +static pthread_barrier_t barrier;
> +
> +static void *
> +thread_func (void *v)
> +{
> +  int i;
> +
> +  pthread_barrier_wait (&barrier);
> +
> +  for (i = 0; i < 100; i++)
> +    sleep (1);
> +}
> +
> +static void
> +all_threads_created (void)
> +{
> +}
> +
> +int
> +main(void)
> +{
> +  int i;
> +  pthread_t threads[NUM_THREADS];
> +
> +  /* +1 to account for the main thread */
> +  pthread_barrier_init (&barrier, NULL, NUM_THREADS + 1);
> +
> +  for (i = 0; i < NUM_THREADS; i++)
> +    pthread_create (&threads[i], NULL, thread_func, NULL);
> +
> +  pthread_barrier_wait (&barrier);
> +
> +  all_threads_created ();
> +
> +  for (i = 0; i < 100; i++)
> +    sleep (1);
> +
> +  return 0;
> +}
> diff --git a/gdb/testsuite/gdb.mi/mi-threads-interrupt.exp b/gdb/testsuite/gdb.mi/mi-threads-interrupt.exp
> new file mode 100644
> index 0000000..af5d99b
> --- /dev/null
> +++ b/gdb/testsuite/gdb.mi/mi-threads-interrupt.exp
> @@ -0,0 +1,75 @@
> +# Copyright 2016 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/>.
> +
> +load_lib mi-support.exp
> +set MIFLAGS "-i=mi"
> +
> +standard_testfile
> +
> +if {[gdb_compile_pthreads "$srcdir/$subdir/$srcfile" $binfile executable {debug}] != "" } {
> +    return -1
> +}
> +
> +# This tests the resolution of PR 20039.
> +#
> +# With a multi-threaded inferior and with MI/all-stop, it was not possible
> +# to interrupt it with ctrl-C after a continue.
> +
> +proc test_continue_interrupt { } {
> +  global binfile
> +  global async
> +
> +  gdb_exit
> +  if {[mi_gdb_start]} {
> +      continue
> +  }
> +
> +  # Load the binary in gdb...
> +  mi_gdb_load $binfile
> +
> +  # ... and run it.
> +  #
> +  # Note this test relies on mi_runto deleting the breakpoint.
> +  # A step-over here would mask the bug.
> +  mi_runto "all_threads_created"
> +
> +  # Consistency check.
> +  mi_check_thread_states {"stopped" "stopped" "stopped"} "check thread states"
> +
> +  # Continue.
> +  mi_gdb_test "565-exec-continue" "565\\^running\r\n\\*running,thread-id=\"all\"" "continue"

I was about to push this, but decided to run the test again.  For some reason, this line
gives me a FAIL, which I didn't see before:

  FAIL: gdb.mi/mi-threads-interrupt.exp: continue

and I can't see what's wrong with:

101 Expecting: ^(.*[^M
102 ]+)?(565\^running^M
103 \*running,thread-id="all"[^M
104 ]+[(]gdb[)] ^M
105 [ ]*)
106 565-exec-continue^M
107 565^running^M
108 *running,thread-id="all"^M
109 =thread-selected,id="3"^M
110 (gdb) ^M
111 FAIL: gdb.mi/mi-threads-interrupt.exp: continue

Changing this line to be

  mi_send_resuming_command "exec-continue" "continue"

makes it work, although I don't see why.  Does it sound like a good change?

> +  if { $async } {
> +    # Workaround for PR 20045, where -exec-continue will output two (gdb) prompts
> +    # when in "mi-async on" mode.  This mi_gdb_test will consume the extra (gdb).
> +    # Once that bug is fixed, this can be removed.
> +    mi_gdb_test
> +  }

I'll remove this, since I'll put the fix for 20045 before this patch.

Re: [PATCH v2] Add mi-threads-interrupt.exp test (PR 20039)

[Simon Marchi]

On 16-05-18 09:44 AM, Pedro Alves wrote:
> On 05/18/2016 02:27 PM, Simon Marchi wrote:
> 
>>> +  # Continue.
>>> +  mi_gdb_test "565-exec-continue" "565\\^running\r\n\\*running,thread-id=\"all\"" "continue"
>>
>> I was about to push this, but decided to run the test again.  For some reason, this line
>> gives me a FAIL, which I didn't see before:
>>
>>   FAIL: gdb.mi/mi-threads-interrupt.exp: continue
>>
>> and I can't see what's wrong with:
>>
>> 101 Expecting: ^(.*[^M
>> 102 ]+)?(565\^running^M
>> 103 \*running,thread-id="all"[^M
>> 104 ]+[(]gdb[)] ^M
>> 105 [ ]*)
>> 106 565-exec-continue^M
>> 107 565^running^M
>> 108 *running,thread-id="all"^M
>> 109 =thread-selected,id="3"^M
>> 110 (gdb) ^M
>> 111 FAIL: gdb.mi/mi-threads-interrupt.exp: continue
>>
>> Changing this line to be
>>
>>   mi_send_resuming_command "exec-continue" "continue"
>>
>> makes it work, although I don't see why.  
> 
> The regex shown above, from mi_gdb_test:
> 
> ...
>  > 103 \*running,thread-id="all"[^M
>  > 104 ]+[(]gdb[)] ^M
> ...
> 
> expects that nothing appears between the "*running" line
> and the prompt.
> 
> Before the fix for PR 20045 (the spurious prompt),
> you probably had:
> 
>  565-exec-continue^M
>  565^running^M
>  *running,thread-id="all"^M
>  (gdb) ^M                        <<<<<<<<< spurious prompt
>  =thread-selected,id="3"^M
>  (gdb) ^M
> 
> which made the mi_gdb_test match output until the first
> prompt.  That is, as if the output was:
> 
>  565-exec-continue^M
>  565^running^M
>  *running,thread-id="all"^M
>  (gdb) ^M
> 
> But now without the spurious prompt we have:
> 
>  565-exec-continue^M
>  565^running^M
>  *running,thread-id="all"^M
>  =thread-selected,id="3"^M
>  (gdb) ^M
> 
> which no longer matches.
> 
> Does it sound like a good change?
> 
> It does.

Ahh it all makes sense now.  For some reason I thought we accepted arbitrary stuff between
the specified regex and the (gdb).

>>
>>> +  if { $async } {
>>> +    # Workaround for PR 20045, where -exec-continue will output two (gdb) prompts
>>> +    # when in "mi-async on" mode.  This mi_gdb_test will consume the extra (gdb).
>>> +    # Once that bug is fixed, this can be removed.
>>> +    mi_gdb_test
>>> +  }
>>
>> I'll remove this, since I'll put the fix for 20045 before this patch.
>>
> 
> BTW, noticed a missing space-before-parens:
> 
>>> +int
>>> +main(void)

Thanks, fixed.

I pushed this final version to both master and gdb-7.11-branch:


From 9e8f9b05add4517189d7724ff3ed7c16f7b04daf Mon Sep 17 00:00:00 2001
From: Simon Marchi <simon.marchi@ericsson.com>
Date: Wed, 18 May 2016 10:13:12 -0400
Subject: [PATCH] Add mi-threads-interrupt.exp test (PR 20039)

Add a new test for PR 20039.  The test spawns new threads, then tries to
interrupt, continue, and interrupt again.  This use case was fixed by
commit 5fe966540d6b748f825774868463003700f0c878 in master, but gdb 7.11
is affected (so if you try it on the gdb-7.11-branch right now, the test
will fail).

New in v2, the test now handles mi-async on mode properly.  The failure
was specific to mi-async off, but I don't think it's bad to test the
same thing under async on mode.  I added a little hack when running in
async mode to work around bug 20045.

I also removed one continue/interrupt pair, as a single one was enough to
trigger the problem.

gdb/testsuite/ChangeLog:

	* gdb.mi/mi-threads-interrupt.c: New file.
	* gdb.mi/mi-threads-interrupt.exp: New file.
---
 gdb/testsuite/ChangeLog                       |  5 ++
 gdb/testsuite/gdb.mi/mi-threads-interrupt.c   | 61 +++++++++++++++++++++++
 gdb/testsuite/gdb.mi/mi-threads-interrupt.exp | 69 +++++++++++++++++++++++++++
 3 files changed, 135 insertions(+)
 create mode 100644 gdb/testsuite/gdb.mi/mi-threads-interrupt.c
 create mode 100644 gdb/testsuite/gdb.mi/mi-threads-interrupt.exp

diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index 6f93044..a773c63 100644
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,3 +1,8 @@
+2016-05-18  Simon Marchi  <simon.marchi@ericsson.com>
+
+	* gdb.mi/mi-threads-interrupt.c: New file.
+	* gdb.mi/mi-threads-interrupt.exp: New file.
+
 2016-05-17  Simon Marchi  <simon.marchi@ericsson.com>

 	PR gdb/18077
diff --git a/gdb/testsuite/gdb.mi/mi-threads-interrupt.c b/gdb/testsuite/gdb.mi/mi-threads-interrupt.c
new file mode 100644
index 0000000..fd69321
--- /dev/null
+++ b/gdb/testsuite/gdb.mi/mi-threads-interrupt.c
@@ -0,0 +1,61 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2016 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/>.  */
+
+#include <pthread.h>
+#include <unistd.h>
+
+#define NUM_THREADS 2
+
+static pthread_barrier_t barrier;
+
+static void *
+thread_func (void *v)
+{
+  int i;
+
+  pthread_barrier_wait (&barrier);
+
+  for (i = 0; i < 100; i++)
+    sleep (1);
+}
+
+static void
+all_threads_created (void)
+{
+}
+
+int
+main (void)
+{
+  int i;
+  pthread_t threads[NUM_THREADS];
+
+  /* +1 to account for the main thread */
+  pthread_barrier_init (&barrier, NULL, NUM_THREADS + 1);
+
+  for (i = 0; i < NUM_THREADS; i++)
+    pthread_create (&threads[i], NULL, thread_func, NULL);
+
+  pthread_barrier_wait (&barrier);
+
+  all_threads_created ();
+
+  for (i = 0; i < 100; i++)
+    sleep (1);
+
+  return 0;
+}
diff --git a/gdb/testsuite/gdb.mi/mi-threads-interrupt.exp b/gdb/testsuite/gdb.mi/mi-threads-interrupt.exp
new file mode 100644
index 0000000..40e98a5
--- /dev/null
+++ b/gdb/testsuite/gdb.mi/mi-threads-interrupt.exp
@@ -0,0 +1,69 @@
+# Copyright 2016 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/>.
+
+load_lib mi-support.exp
+set MIFLAGS "-i=mi"
+
+standard_testfile
+
+if {[gdb_compile_pthreads "$srcdir/$subdir/$srcfile" $binfile executable {debug}] != "" } {
+    return -1
+}
+
+# This tests the resolution of PR 20039.
+#
+# With a multi-threaded inferior and with MI/all-stop, it was not possible
+# to interrupt it with ctrl-C after a continue.
+
+proc test_continue_interrupt { } {
+  global binfile
+  global async
+
+  gdb_exit
+  if {[mi_gdb_start]} {
+      continue
+  }
+
+  # Load the binary in gdb...
+  mi_gdb_load $binfile
+
+  # ... and run it.
+  #
+  # Note this test relies on mi_runto deleting the breakpoint.
+  # A step-over here would mask the bug.
+  mi_runto "all_threads_created"
+
+  # Consistency check.
+  mi_check_thread_states {"stopped" "stopped" "stopped"} "check thread states"
+
+  # Continue.
+  mi_send_resuming_command "exec-continue" "continue"
+
+  # Wait a bit to make sure all MI events are sent, before sending the
+  # interruption request.
+  sleep 1
+
+  # Send the interrupt request.
+  if { $async } {
+    mi_gdb_test "888-exec-interrupt" "888\\^done" "interrupt"
+  } else {
+    send_gdb "\003"
+  }
+
+  # Wait for the *stopped.
+  mi_expect_interrupt "interrupt reported"
+}
+
+test_continue_interrupt
-- 
2.8.2