On Fri, Jan 25, 2019 at 12:21 PM Palmer Dabbelt <[email protected]> wrote:
> Jim, would you be opposed to something like this?
This looks OK to me.
> + builtin_define_with_int_value ("__riscv_tune_misaligned_load_cost",
> +
> riscv_tune_info->slow_unaligned_access ? 1024 : 1);
> + builtin_define_with_int_value ("__riscv_tune_misaligned_store_cost",
> +
> riscv_tune_info->slow_unaligned_access ? 1024 : 1);
It would be nice to have a better way to compute these values, maybe
an extra field in the tune structure, but we can always worry about
that later when we need it.
Jim