Re: RFA symtab: Fix for PR c++/1267 ("next" and shared libraries)

[Daniel Jacobowitz <drow@mvista.com>]

On Mon, Jul 21, 2003 at 02:22:34PM -0400, Andrew Cagney wrote:
> >
> >No, really, I still think that the comment is wrong.  The get_frame_id
> >call triggers the prologue analyzer to analyze the NEXT frame.  I.E.
> >this_id is called with the NEXT frame, and THIS cache.
> 
> Nope. get_frame_id triggers the analysis of THIS frame's prologue (i.e., 
> next_frame & this_cache).
> 
> >The call to frame_pc_unwind uses THIS frame and creates the PREV cache. 
> >It's one frame off on the stack.  The frame it would be saving from
> >prologue analysis is not the one we'd need to analyze for the
> >get_frame_id call.
> 
> Nope.  frame_pc_unwind would turn into frame_register_unwind (this, 
> PC_REGNUM) and that will also trigger the analysis of THIS frame's 
> prologue (i.e., next_frame & this_cache).

Ah, I see where I was confused.  Thanks.  I really need to spend more
time looking at this.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer