Re: PR target/8869

[Andrew Pinski <pinskia at physics dot uc dot edu>]

Is this really fixing PR 7277, not 8869 which is meta-bug for two bugs, 
7277 and 10984?
Or does it fix both of them?

Thanks,
Andrew Pinski

On Tuesday, Aug 19, 2003, at 10:02 US/Eastern, Jan Hubicka wrote:

> Tue Aug 19 16:01:36 CEST 2003  Jan Hubicka  <jh@suse.cz>
> 	PR target/8869
> 	* expr.c (convert_modes): Deal properly with integer to vector
> 	constant conversion.
> Index: expr.c
> ===================================================================
> RCS file: /cvs/gcc/gcc/gcc/expr.c,v
> retrieving revision 1.576
> diff -c -3 -p -r1.576 expr.c
> *** expr.c	12 Aug 2003 01:45:22 -0000	1.576
> --- expr.c	19 Aug 2003 14:01:14 -0000
> *************** convert_modes (enum machine_mode mode, e
> *** 1419,1424 ****
> --- 1419,1433 ----
>         return gen_lowpart (mode, x);
>       }
>
> +   /* Converting from integer constant into mode is always equivalent 
> to an
> +      subreg operation.  */
> +   if (VECTOR_MODE_P (mode) && GET_MODE (x) == VOIDmode)
> +     {
> +       if (GET_MODE_BITSIZE (mode) != GET_MODE_BITSIZE (oldmode))
> + 	abort ();
> +       return simplify_gen_subreg (mode, x, oldmode, 0);
> +     }
> +
>     temp = gen_reg_rtx (mode);
>     convert_move (temp, x, unsignedp);
>     return temp;