Re: [RFA] dwarf2_physname

[Tom Tromey <tromey@redhat.com>]

>>>>> "Keith" == Keith Seitz <keiths@redhat.com> writes:

Keith> In an effort to help maintainers assess the code in smaller chunks
Keith> without introducing new, temporary test suite failures, I'm going to
Keith> try something "different" -- this patch, and the follow-ons, will not
Keith> actually do anything until a final patch turns it all on. My hope is
Keith> that this will be minimally intrusive to GDB's code base, especially
Keith> given all the talk of a new release.

This seems fine to me, though as a procedural issue I would suggest that
you not commit anything until the whole series is approved.  (The
alternative seems to be committing unused code, which I think we should
not do in general.)

If this is a hardship for some reason we can discuss it.


I haven't read your patch in depth yet but I wanted to mention a couple
of things.  I'll look at the rest soon.

Keith> 	Based on work from Daniel Jacobowitz  <dan@codesourcery.com>:
Keith> 	* typeprint.h (c_type_print_args): Declare.
Keith> 	* ui-file.h (ui_file_obsavestring): Declare.
Keith> 	* c-typeprint.c (c_type_print_args): Add to global namespace.
Keith> 	Add "show_articifical" parameter and only print them when
Keith> 	requested.
Keith> 	* ui-file.c (do_ui_file_obsavestring): New function.
Keith> 	(ui_file_obsavestring): New function.

All of this part looked ok.

Keith> +/* A vector used during linkage name generation.  */
Keith> +typedef struct die_info *die_info_p;
Keith> +DEF_VEC_P (die_info_p);
Keith> +static VEC(die_info_p) *die_list;

I wonder if a global is really necessary.
I didn't check.

Keith> +      VEC_quick_push (die_info_p, die_list, d);

AFAICT the vector is only added to with VEC_quick_push.
This will die, though, if there is insufficient space in the vec.

So, something must be missing here -- either an explicit initialization
of the vec, or this should use VEC_safe_push.

Tom