On Wed, 16 Apr 2025 at 15:17, Tomasz Kaminski <tkami...@redhat.com> wrote:
>
>
>
> On Wed, Apr 16, 2025 at 3:47 PM Tomasz Kamiński <tkami...@redhat.com> wrote:
>>
>> Because the _M_format(__rg, __fc) were placed outside of if constexpr,
>> these method and it's childs where instantiated, even if _M_format<const
>> _Range>
>> could be used. Now we put the calls in else branch of if constexpr.
>>
>> libstdc++-v3/ChangeLog:
>>
>> * include/std/format (range_formatter::format): Do not instantiate
>> _M_format for mutable _Rg if const _Rg can be used.
>> ---
>> Testing on x86_64-linux, but there are no tests that will detect
>> uncessary instantiations.
>
> This is not a correctness issue, only some dead code elimination, so
> does not need to be in v15.
OK for trunk. If that's pushed before gcc-15 branches, great.
>>
>>
>> libstdc++-v3/include/std/format | 15 +++++++++------
>> 1 file changed, 9 insertions(+), 6 deletions(-)
>>
>> diff --git a/libstdc++-v3/include/std/format
>> b/libstdc++-v3/include/std/format
>> index 27253f50ea8..d44c4b24abe 100644
>> --- a/libstdc++-v3/include/std/format
>> +++ b/libstdc++-v3/include/std/format
>> @@ -5636,12 +5636,15 @@ namespace __format
>> {
>> using _Range = remove_reference_t<_Rg>;
>> if constexpr (__format::__const_formattable_range<_Range, _CharT>)
>> - {
>> - using _CRef = ranges::range_reference_t<const _Range>;
>> - if constexpr (same_as<remove_cvref_t<_CRef>, _Tp>)
>> - return _M_format<const _Range>(__rg, __fc);
>> - }
>> - return _M_format(__rg, __fc);
>> + {
>> + using _CRef = ranges::range_reference_t<const _Range>;
>> + if constexpr (same_as<remove_cvref_t<_CRef>, _Tp>)
>> + return _M_format<const _Range>(__rg, __fc);
>> + else
>> + return _M_format(__rg, __fc);
>> + }
>> + else
>> + return _M_format(__rg, __fc);
>> }
>>
>> private:
>> --
>> 2.49.0
>>
