Re: [patch] Fix internal error on store to FP component at -O2

[Richard Biener via Gcc-patches]

On Mon, May 25, 2020 at 2:14 PM Eric Botcazou <botca...@adacore.com> wrote:
>
> > Hmm, MAX_BITSIZE_MODE_ANY_INT looks a bit arbitrary since on
> > x86 it (IIRC) includes things like OImode. Maybe MOVE_MAX or
> > MAX_FIXED_MODE_SIZE are better suited here?
>
> I forgot to mention that I picked MAX_BITSIZE_MODE_ANY_INT because of:
>
>   bool invalid = (base_addr == NULL_TREE
>                   || (maybe_gt (bitsize,
>                                 (unsigned int) MAX_BITSIZE_MODE_ANY_INT)
>                       && TREE_CODE (rhs) != INTEGER_CST
>                       && (TREE_CODE (rhs) != CONSTRUCTOR
>                           || CONSTRUCTOR_NELTS (rhs) != 0)));

That's likely because of the encoding limits of wide_ints I guess?  That is,
an implementation limit rather than one for code-quality.

Richard.

> --
> Eric Botcazou