Re: [patch] segfault in get_prev_frame

[Andrew Cagney <cagney@gnu.org>]

Randolph Chung wrote:
> sigh, found a bug in the inside_entry_func logic that i just
> committed...
> 
> if you jump backwards in a backtrace (i.e you are at frame 0, and you
> do "frame 4") you will get a segfault because the frame unwinder is not
> yet filled in. using get_frame_type will do the necessary lazy
> initialization.
> 
> ok to check in?

Yes (add it to your raincheck :-).

Andrew

> 2004-11-10  Randolph Chung  <tausq@debian.org>
> 
> 	* frame.c (get_prev_frame): Use get_frame_type to retrieve the frame's
> 	type.
> 
> Index: frame.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/frame.c,v
> retrieving revision 1.194
> diff -u -p -r1.194 frame.c
> --- frame.c	10 Nov 2004 21:17:56 -0000	1.194
> +++ frame.c	10 Nov 2004 22:41:40 -0000
> @@ -1239,7 +1239,7 @@ get_prev_frame (struct frame_info *this_
>       stop at main, we should at least stop at the entry point of the
>       application.  */
>    if (!backtrace_past_entry
> -      && this_frame->unwind->type != DUMMY_FRAME && this_frame->level >= 0
> +      && get_frame_type (this_frame) != DUMMY_FRAME && this_frame->level >= 0
>        && inside_entry_func (this_frame))
>      {
>        frame_debug_got_null_frame (gdb_stdlog, this_frame, "inside entry func");
>