Re: [rfa:symtab] deprecate inside_entry_func

[Kevin Buettner <kevinb at redhat dot com>]

On Oct 31,  7:46pm, Andrew Cagney wrote:

> > On Oct 31,  7:07pm, Andrew Cagney wrote:
> > 
> >> This patch deprecates the function inside_entry_func.  Per the new comments:
> >> 
> >> +  /* NOTE: cagney/2003-10-31: A very simple test, such as
> >> +     get_frame_func == entry_point should be sufficient for
> >> +     identifying a pc in the entry function.  Does anyone know why it
> >> +     wasn't sufficient and hence, why the very convoluted
> >> +     "deprecated_inside_entry_func" is needed.  */
> >> +  /* NOTE: cagney/2003-10-31: An ABI and its crt0 code should define
> >> +     and implement a clean frame termination.  Not doing that is
> >> +     really a bug in the ABI/crt0, and, hence, not a reason for
> >> +     enabling the call to deprecated_inside_entry_func.  */
> > 
> > I do agree that it'd be nice if all ABIs provided a clean way to detect
> > the bottom-most frame.  Not all of them do however, and in such cases,
> > a mechanism like inside_entry_func() is necessary.
> 
> > So, in short, I oppose the deprecation of this function.
> 
> Kevin, you previously wrote:
> 
> >> I'd like to avoid re-introducing a dependency on inside_entry_func() as 
> >> that places garish requirements on the object file readers :-(
> > 
> > I agree that object file readers should not attempt to track of
> > the bounds of the start function.  However, given an arbitrary
> > address, it's not unreasonable to ask the symtab machinery to attempt
> > to figure out the function bounds.  And, in fact, this is just what
> > find_pc_partial_function() does.

Yes, the reason I wrote this was to note that there are other ways of
implementing inside_entry_func() which wouldn't place garish
requirements on the object file readers.

Kevin