[PATCH v25 25/33] libstdc++: Optimize std::is_function compilation performance

Ken Matsui kmatsui@cs.washington.edu
Thu Dec 7 03:20:24 GMT 2023


On Tue, Oct 24, 2023 at 4:02 AM Jonathan Wakely <jwakely@redhat.com> wrote:
>
>
>
> On Tue, 24 Oct 2023 at 03:16, Ken Matsui <kmatsui@gcc.gnu.org> wrote:
>>
>> This patch optimizes the compilation performance of std::is_function
>> by dispatching to the new __is_function built-in trait.
>>
>> libstdc++-v3/ChangeLog:
>>
>>         * include/std/type_traits (is_function): Use __is_function
>>         built-in trait.
>>         (is_function_v): Likewise. Optimize its implementation.
>>         (is_const_v): Move on top of is_function_v as is_function_v now
>>         depends on is_const_v.
>
>
> I think I'd prefer to keep is_const_v where it is now, adjacent to is_volatile_v, and move is_function_v after those.
>
> i.e. like this (but with the additional changes to use the new built-in):
>
> --- a/libstdc++-v3/include/std/type_traits
> +++ b/libstdc++-v3/include/std/type_traits
> @@ -3198,8 +3198,8 @@ template <typename _Tp>
>    inline constexpr bool is_union_v = __is_union(_Tp);
>  template <typename _Tp>
>    inline constexpr bool is_class_v = __is_class(_Tp);
> -template <typename _Tp>
> -  inline constexpr bool is_function_v = is_function<_Tp>::value;
> +// is_function_v is defined below, after is_const_v.
> +
>  template <typename _Tp>
>    inline constexpr bool is_reference_v = false;
>  template <typename _Tp>
> @@ -3226,6 +3226,8 @@ template <typename _Tp>
>    inline constexpr bool is_volatile_v = false;
>  template <typename _Tp>
>    inline constexpr bool is_volatile_v<volatile _Tp> = true;
> +template <typename _Tp>
> +  inline constexpr bool is_function_v = is_function<_Tp>::value;
>
>  template <typename _Tp>
>    inline constexpr bool is_trivial_v = __is_trivial(_Tp);
>
> The variable templates are currently defined in the order shown in the standard, in te [meta.type.synop] synopsis, and in the [meta.unary.cat] table. So let's move _is_function_v later and add a comment saying why it's not in the expected place.
>
>

Sorry for missing this.  Thank you for your review!  That totally
makes sense, and I will update my patches accordingly.


More information about the Libstdc++ mailing list