On Fri, Apr 29, 2016 at 2:34 PM, Richard Sandiford
<richard.sandif...@arm.com> wrote:
> While looking at the use of cst_and_fits_in_hwi in tree-ssa-loop-ivopts.c,
> I had difficulty working out what the function actually tests. The
> final NUNITS check seems redundant, since it asks about the number of
> HWIs in the _unextended_ constant. We've already checked that the
> unextended constant has no more than HOST_BITS_PER_WIDE_INT bits, so the
> length must be 1.
>
> I think this was my fault, sorry.
>
> Tested on x86_64-linux-gnu and aarch64-linux-gnu. OK to install?
Ok.
Richard.
> Thanks,
> Richard
>
>
> gcc/
> * tree.c (cst_and_fits_in_hwi): Simplify.
>
> Index: gcc/tree.c
> ===================================================================
> --- gcc/tree.c
> +++ gcc/tree.c
> @@ -1675,13 +1675,8 @@ build_low_bits_mask (tree type, unsigned bits)
> bool
> cst_and_fits_in_hwi (const_tree x)
> {
> - if (TREE_CODE (x) != INTEGER_CST)
> - return false;
> -
> - if (TYPE_PRECISION (TREE_TYPE (x)) > HOST_BITS_PER_WIDE_INT)
> - return false;
> -
> - return TREE_INT_CST_NUNITS (x) == 1;
> + return (TREE_CODE (x) == INTEGER_CST
> + && TYPE_PRECISION (TREE_TYPE (x)) <= HOST_BITS_PER_WIDE_INT);
> }
>
> /* Build a newly constructed VECTOR_CST node of length LEN. */
