This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RS6000] Fix PR12817
On Nov 27, 2004, at 2:30 PM, Alan Modra wrote:
On Sat, Nov 27, 2004 at 02:21:30PM -0800, Dale Johannesen wrote:
+ info_ptr->world_save_p = 1;
+ info_ptr->world_save_p
+ = (WORLD_SAVE_P (info_ptr)
+ && DEFAULT_ABI == ABI_DARWIN
+ && ! (current_function_calls_setjmp && flag_exceptions)
+ && info_ptr->first_fp_reg_save == FIRST_SAVED_FP_REGNO
+ && info_ptr->first_gp_reg_save == FIRST_SAVED_GP_REGNO
+ && info_ptr->first_altivec_reg_save ==
FIRST_SAVED_ALTIVEC_REGNO
+ && info_ptr->cr_save_p);
The first line is redundant.
Note the WORLD_SAVE_P test on the next line. It's part of the trick
to
have this function reduce down to
info_ptr->world_save_p = 0;
on anything but Darwin.
That should happen whether there's a previous assignment to 1 or not,
surely?
Oh, I see, it hadn't been set previously on Darwin. OK then.