A few simple DImode improvements

[Bernd Schmidt]

On 06/29/2010 01:03 AM, Ian Lance Taylor wrote:
> I see what you're getting at.  It's unlikely, but it does happen in code
> which examines the bits of a floating point number via a union or using
> -fno-strict-aliasing and a type cast.  So if that code works correctly
> if perhaps less efficiently, then I agree that checking that the mode
> class is MODE_INT should work OK here.

Hmm.  Looking at this again now, aren't we already doing that?
(resolve_reg_p checks whether DEST is a CONCAT)

  if (!can_decompose_p (dest)
      || (side_effects_p (dest) && !pushing)
      || (!SCALAR_INT_MODE_P (dest_mode)
	  && !resolve_reg_p (dest)
	  && !resolve_subreg_p (dest)))


Bernd