Re: 23.0.50; Emacs seg fault

[storm@cua.dk (Kim F. Storm)]

Leo <sdl.web@gmail.com> writes:

> #1  0x08194676 in wait_reading_process_output (time_limit=30, microsecs=0, read_kbd=-1, do_display=1, wait_for_cell=137551137, 
>     wait_proc=0x0, just_wait_proc=0) at process.c:4880

I doubt that is the right line number due to compiler optimizations.
Compile with CFLAGS="-g -O0" to get max debugging info.

Since xerrno = 9  (EBADF), I suspect it actually traps in
process.c line 4651 

This means that one of the bits in the Available mask does not
correspond to a valid file descriptor.

Looking at the code, I see that this piece of code is executed
between setting up the Available mask and the call to select.

      /* If frame size has changed or the window is newly mapped,
	 redisplay now, before we start to wait.  There is a race
	 condition here; if a SIGIO arrives between now and the select
	 and indicates that a frame is trashed, the select may block
	 displaying a trashed screen.  */
      if (frame_garbaged && do_display)
	{
	  clear_waiting_for_input ();
	  redisplay_preserve_echo_area (11);
	  if (read_kbd < 0)
	    set_waiting_for_input (&timeout);
	}

This looks dangerous to me.

Could you try this patch and see if it fixes the problem:

*** process.c	20 Sep 2007 12:50:59 +0200	1.520
--- process.c	30 Sep 2007 12:45:50 +0200	
***************
*** 4509,4514 ****
--- 4509,4527 ----
  	  break;
  	}
  
+       /* If frame size has changed or the window is newly mapped,
+ 	 redisplay now, before we start to wait.  There is a race
+ 	 condition here; if a SIGIO arrives between now and the select
+ 	 and indicates that a frame is trashed, the select may block
+ 	 displaying a trashed screen.  */
+       if (frame_garbaged && do_display)
+ 	{
+ 	  clear_waiting_for_input ();
+ 	  redisplay_preserve_echo_area (11);
+ 	  if (read_kbd < 0)
+ 	    set_waiting_for_input (&timeout);
+ 	}
+ 
        /* Wait till there is something to do */
  
        if (wait_proc && just_wait_proc)
***************
*** 4535,4553 ****
   	  check_delay = wait_channel >= 0 ? 0 : process_output_delay_count;
  	}
  
-       /* If frame size has changed or the window is newly mapped,
- 	 redisplay now, before we start to wait.  There is a race
- 	 condition here; if a SIGIO arrives between now and the select
- 	 and indicates that a frame is trashed, the select may block
- 	 displaying a trashed screen.  */
-       if (frame_garbaged && do_display)
- 	{
- 	  clear_waiting_for_input ();
- 	  redisplay_preserve_echo_area (11);
- 	  if (read_kbd < 0)
- 	    set_waiting_for_input (&timeout);
- 	}
- 
        no_avail = 0;
        if (read_kbd && detect_input_pending ())
  	{
--- 4548,4553 ----


-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk