This is the mail archive of the
mailing list for the GCC project.
Re: More on addressof: how to fix the backends that use them?
- From: Richard Henderson <rth at redhat dot com>
- To: Steven Bosscher <stevenb at suse dot de>
- Cc: gcc at gcc dot gnu dot org, jh at suse dot cz
- Date: Sun, 6 Jun 2004 13:15:02 -0700
- Subject: Re: More on addressof: how to fix the backends that use them?
- References: <firstname.lastname@example.org>
On Tue, Jun 01, 2004 at 11:34:13PM +0200, Steven Bosscher wrote:
> alpha_gp_save_rtx (void)
> rtx r = get_hard_reg_initial_val (DImode, 29);
> if (GET_CODE (r) != MEM)
> r = gen_mem_addressof (r, NULL_TREE, /*rescan=*/true);
> return r;
> What can this code be replaced with such that it does not
> create an ADDRESSOF?
Use a memory. I don't much care if systems with the buggy ld
get a teeny bit slower.