gdb/inflow.c: Move SIGTTOU temporary ignoring to a RAII class

  I expect to use this more places (in inflow.c) in follow up patches,
but I think this is still good on its own.

gdb/ChangeLog:
yyyy-mm-dd  Pedro Alves  <palves@redhat.com>

	* inflow.c (scoped_ignore_sigttou): New class.
	(child_terminal_ours_1, new_tty): Use it.
---
 gdb/inflow.c | 50 +++++++++++++++++++++++++++++++++-----------------
 1 file changed, 33 insertions(+), 17 deletions(-)
  

On 9 November 2017 19:04:04 GMT-05:00, Pedro Alves <palves@redhat.com> wrote:
>I expect to use this more places (in inflow.c) in follow up patches,
>but I think this is still good on its own.
>
>gdb/ChangeLog:
>yyyy-mm-dd  Pedro Alves  <palves@redhat.com>
>
>	* inflow.c (scoped_ignore_sigttou): New class.
>	(child_terminal_ours_1, new_tty): Use it.
>---
> gdb/inflow.c | 50 +++++++++++++++++++++++++++++++++-----------------
> 1 file changed, 33 insertions(+), 17 deletions(-)
>
>diff --git a/gdb/inflow.c b/gdb/inflow.c
>index 2fba0fa..52be5af 100644
>--- a/gdb/inflow.c
>+++ b/gdb/inflow.c
>@@ -91,6 +91,35 @@ static serial_ttystate initial_gdb_ttystate;
> 
>static struct terminal_info *get_inflow_inferior_data (struct inferior
>*);
> 
>+/* RAII class used to ignore SIGTTOU in a scope.  */
>+
>+class scoped_ignore_sigttou
>+{
>+public:
>+  scoped_ignore_sigttou ()
>+  {
>+#ifdef SIGTTOU
>+    if (job_control)
>+      m_osigttou = signal (SIGTTOU, SIG_IGN);
>+#endif
>+  }
>+
>+  ~scoped_ignore_sigttou ()
>+  {
>+#ifdef SIGTTOU
>+    if (job_control)
>+      signal (SIGTTOU, m_osigttou);
>+#endif
>+  }
>+
>+  DISABLE_COPY_AND_ASSIGN (scoped_ignore_sigttou);
>+
>+private:
>+#ifdef SIGTTOU
>+  sighandler_t m_osigttou = NULL;
>+#endif
>+};
>+
> #ifdef HAVE_TERMIOS_H
> 
> /* Return the process group of the current inferior.  */
>@@ -329,17 +358,11 @@ child_terminal_ours_1 (int output_only)
>     return;
>   else
>     {
>-#ifdef SIGTTOU
>-      /* Ignore this signal since it will happen when we try to set
>the
>-         pgrp.  */
>-      sighandler_t osigttou = NULL;
>-#endif
>       int result ATTRIBUTE_UNUSED;
> 
>-#ifdef SIGTTOU
>-      if (job_control)
>-	osigttou = signal (SIGTTOU, SIG_IGN);
>-#endif
>+      /* Ignore SIGTTOU since it will happen when we try to set the
>+	 terminal's pgrp.  */
>+      scoped_ignore_sigttou ignore_sigttou;
> 
>       xfree (tinfo->ttystate);
>       tinfo->ttystate = serial_get_tty_state (stdin_serial);
>@@ -372,11 +395,6 @@ child_terminal_ours_1 (int output_only)
> #endif /* termios */
> 	}
> 
>-#ifdef SIGTTOU
>-      if (job_control)
>-	signal (SIGTTOU, osigttou);
>-#endif
>-
>       if (!job_control)
> 	{
> 	  signal (SIGINT, sigint_ours);
>@@ -603,12 +621,10 @@ new_tty (void)
>   tty = open ("/dev/tty", O_RDWR);
>   if (tty > 0)
>     {
>-      sighandler_t osigttou;
>+      scoped_ignore_sigttou ignore_sigttou;
> 
>-      osigttou = signal (SIGTTOU, SIG_IGN);
>       ioctl (tty, TIOCNOTTY, 0);
>       close (tty);
>-      signal (SIGTTOU, osigttou);
>     }
> #endif
> 

LGTM (at least on my phone :))

Simon
  

On 11/10/2017 04:24 PM, Simon Marchi wrote:
> On 9 November 2017 19:04:04 GMT-05:00, Pedro Alves <palves@redhat.com> wrote:
>> I expect to use this more places (in inflow.c) in follow up patches,
>> but I think this is still good on its own.


> LGTM (at least on my phone :))

:-)

Now pushed.

Thanks,
Pedro Alves