[v3,4/7] Test case for Inferior.thread_from_thread_handle

  As the title says, this is a test case for
Inferior.thread_from_thread_handle, a python method which will,
given a thread library dependent thread handle, find the GDB thread
which corresponds to that thread handle (in the inferior under
consideration).

The C file for this test case causes the thread handles for the
main thread and two child threads to be placed into an array.  The
test case runs to one of the functions (do_something()) at which point,
it retrieves the thread handles from the array and attempts to find the
correponding thread in GDB's internal thread list.

I use barriers to make sure that both threads have actually started;
execution will stop when one of the threads breaks at do_something.

Thanks to Simon Marchi for suggestions for forcing the thread
numbering to be stable.

gdb/testsuite/ChangeLog:
    
    	* gdb.python/py-thrhandle.c, gdb.python/py-thrhandle.exp: New
    	files.
---
 gdb/testsuite/gdb.python/py-thrhandle.c   | 97 +++++++++++++++++++++++++++++++
 gdb/testsuite/gdb.python/py-thrhandle.exp | 72 +++++++++++++++++++++++
 2 files changed, 169 insertions(+)
  

Hi Kevin,

On 2017-07-19 02:55, Kevin Buettner wrote:
> As the title says, this is a test case for
> Inferior.thread_from_thread_handle, a python method which will,
> given a thread library dependent thread handle, find the GDB thread
> which corresponds to that thread handle (in the inferior under
> consideration).
> 
> The C file for this test case causes the thread handles for the
> main thread and two child threads to be placed into an array.  The
> test case runs to one of the functions (do_something()) at which point,
> it retrieves the thread handles from the array and attempts to find the
> correponding thread in GDB's internal thread list.

"correponding"

> 
> I use barriers to make sure that both threads have actually started;
> execution will stop when one of the threads breaks at do_something.
> 
> Thanks to Simon Marchi for suggestions for forcing the thread
> numbering to be stable.
> 
> gdb/testsuite/ChangeLog:
> 
>     	* gdb.python/py-thrhandle.c, gdb.python/py-thrhandle.exp: New
>     	files.
> ---
>  gdb/testsuite/gdb.python/py-thrhandle.c   | 97 
> +++++++++++++++++++++++++++++++
>  gdb/testsuite/gdb.python/py-thrhandle.exp | 72 +++++++++++++++++++++++
>  2 files changed, 169 insertions(+)
> 
> diff --git a/gdb/testsuite/gdb.python/py-thrhandle.c
> b/gdb/testsuite/gdb.python/py-thrhandle.c
> new file mode 100644
> index 0000000..10e0222
> --- /dev/null
> +++ b/gdb/testsuite/gdb.python/py-thrhandle.c
> @@ -0,0 +1,97 @@
> +/* This testcase is part of GDB, the GNU debugger.
> +
> +   Copyright 2017 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>
> +#include <memory.h>
> +
> +#define NTHR 3
> +#define NBOGUSTHR 2
> +
> +int thr_data[NTHR];
> +
> +/* Thread handles for each thread plus some "bogus" threads.  */
> +pthread_t thrs[NTHR + NBOGUSTHR];
> +
> +/* The thread children will meet at this barrier. */
> +pthread_barrier_t c_barrier;
> +
> +/* The main thread and child thread will meet at this barrier. */
> +pthread_barrier_t mc_barrier;
> +
> +void
> +do_something (int n)
> +{
> +}
> +
> +void *
> +do_work (void *data)
> +{
> +  int num = * (int *) data;
> +
> +  /* As the child threads are created, they'll meet the main thread
> +     at this barrier.  We do this to ensure that threads end up in
> +     GDB's thread list in the order in which they were created.  
> Having
> +     this ordering makes it easier to write the test.  */
> +  pthread_barrier_wait (&mc_barrier);
> +
> +  /* All of the child threads will meet at this barrier before 
> proceeding.
> +     This ensures that all threads will be active (not exited) and in
> +     roughly the same state when the first one hits the breakpoint in
> +     do_something().  */
> +  pthread_barrier_wait (&c_barrier);
> +
> +  do_something (num);
> +
> +  pthread_exit (NULL);
> +}
> +
> +void
> +after_mc_barrier (void)
> +{
> +}
> +
> +int
> +main (int argc, char **argv)
> +{
> +  int i;
> +
> +  pthread_barrier_init (&c_barrier, NULL, NTHR - 1);
> +
> +  thrs[0] = pthread_self ();
> +  thr_data[0] = 1;
> +
> +  /* Create two bogus thread handles.  */
> +  memset (&thrs[NTHR], 0, sizeof (pthread_t));
> +  memset (&thrs[NTHR + 1], 0xaa, sizeof (pthread_t));
> +
> +  for (i = 1; i < NTHR; i++)
> +    {
> +      thr_data[i] = i + 1;
> +
> +      pthread_barrier_init (&mc_barrier, NULL, 2);

I don't think it's wrong to init/destroy the barrier repeatedly, but 
it's not necessary.  After the barrier has "opened", it is reset with 
the count it was initialized with.  So you can just call _wait again on 
it.

> +
> +      pthread_create (&thrs[i], NULL, do_work, &thr_data[i]);
> +
> +      pthread_barrier_wait (&mc_barrier);
> +      pthread_barrier_destroy (&mc_barrier);
> +      after_mc_barrier ();
> +    }
> +
> +  for (i = 1; i < NTHR; i++)
> +    pthread_join (thrs[i], NULL);
> +}
> diff --git a/gdb/testsuite/gdb.python/py-thrhandle.exp
> b/gdb/testsuite/gdb.python/py-thrhandle.exp
> new file mode 100644
> index 0000000..033737d
> --- /dev/null
> +++ b/gdb/testsuite/gdb.python/py-thrhandle.exp
> @@ -0,0 +1,72 @@
> +# Copyright (C) 2017 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/>.
> +
> +# Please email any bugs, comments, and/or additions to this file to:
> +# bug-gdb@gnu.org
> +
> +# This file verifies that gdb.thread_from_thread_handle works as 
> expected.

This is now gdb.Inferior.thread_from_thread_handle.

> +
> +standard_testfile
> +
> +
> +if {[gdb_compile_pthreads "${srcdir}/${subdir}/${srcfile}"
> "${binfile}" executable debug] != "" } {
> +    return -1
> +}
> +
> +clean_restart ${binfile}
> +runto_main
> +
> +gdb_test "break after_mc_barrier" \
> +    "Breakpoint 2 at .*: file .*${srcfile}, line .*" \
> +         "breakpoint on after_mc_barrier"
> +
> +gdb_test "break do_something" \
> +    "Breakpoint 3 at .*: file .*${srcfile}, line .*" \
> +         "breakpoint on do_something"
> +
> +gdb_test "continue" \
> +	"Breakpoint 2, after_mc_barrier .*" \
> +	"run to after_mc_barrier (1)"

Do not use parentheses at the end of the test name:
https://sourceware.org/gdb/wiki/GDBTestcaseCookbook#Do_not_use_.22tail_parentheses.22_on_test_messages

Alternatively, you can use gdb_continue_to_breakpoint or gdb_continue 
(not sure why we have both).

> +
> +gdb_test_no_output "del 2" "delete after_mc_barrier breakpoint"
> +
> +gdb_test "continue" \
> +	"Breakpoint 3, do_something .*" \
> +	"run to do_something"
> +
> +# The current thread, indicated by '*' in the "info threads" output
> +# should be stopped in do_something() with a value of n which is the
> +# same as the number reported in the "Id" column.  If it's not, then
> +# something went wrong with the start up sequence which should cause
> +# the main thread to be thread 1, the first child thread to be thread
> +# 2, and the second child thread to be thread 3.
> +
> +gdb_test "info threads"  \
> +	{.*[\r\n]+\* +([0-9]+) +Thread[^\r\n]* do_something \(n=\1\) at.*}

I didn't understand the comment when reading it at first, then wondered 
for way too long how we could end up with n=1 :).

> +gdb_test "python print
> gdb.selected_inferior().thread_from_thread_handle(gdb.parse_and_eval('thrs\[0\]')).num"
> \
> +	"1"
> +
> +gdb_test "python print
> gdb.selected_inferior().thread_from_thread_handle(gdb.parse_and_eval('thrs\[1\]')).num"
> \
> +	"2"
> +
> +gdb_test "python print
> gdb.selected_inferior().thread_from_thread_handle(gdb.parse_and_eval('thrs\[2\]')).num"
> \
> +	"3"
> +
> +gdb_test "python print
> gdb.selected_inferior().thread_from_thread_handle(gdb.parse_and_eval('thrs\[3\]'))"
> \
> +	"None"
> +
> +gdb_test "python print
> gdb.selected_inferior().thread_from_thread_handle(gdb.parse_and_eval('thrs\[4\]'))"
> \
> +	"None"

Please use parentheses with print, so it works with Python 3.  But then, 
your test name will have tail parentheses, so it might be a good idea to 
give them explicit names.

Thanks,

Simon