Remove ira_{zero,one}_hard_reg_set

[Vladimir Makarov]

Richard Sandiford wrote:
> Very minor cleanup, but I noticed that ira_zero_hard_reg_set is
> only ever used in "equal to empty" or "subset of empty" conditions.
> They can be done more directly (and in theory, more efficiently)
> using hard_reg_set_empty_p instead.  ira_one_hard_reg_set isn't
> used at all.
>
> Bootstrapped & regression-tested on x86_64-linux-gnu.  OK to install?
>
>
>   
Ok.  Thanks for finding this, Richard.
> gcc/
> 	* ira-int.h (ira_zero_hard_reg_set, ira_one_hard_reg_set): Delete.
> 	* ira-color.c (setup_allocno_left_conflicts_num): Use
> 	hard_reg_set_equal_p instead of a comparison with
> 	ira_zero_hard_reg_set.
> 	* ira.c (setup_reg_subclasses): Likewise.
> 	(setup_cover_and_important_classes): Likewise.
> 	(setup_class_translate): Likewise.
> 	(setup_reg_class_intersect_union): Likewise.
> 	(ira_zero_hard_reg_set, ira_one_hard_reg_set): Delete.
> 	(ira_init_once): Don't initialize them.
>
>