Date: Wed, 7 Feb 2024 11:25:20 -0500
Thank you, as the next call is next week Feb 14, and it will be focused on
Embedded. I would like to add all these to the agenda. Cheers.
On Mon, Feb 5, 2024 at 4:46 PM Ben Craig via SG14 <sg14_at_[hidden]>
wrote:
> Treating constexpr functions and containers as consteval on freestanding
> is on my roadmap, but I haven't started doing any of the prototyping or
> testing necessary to write a paper.
>
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2020/p2268r0.html#constexpr_to_consteval
>
> I haven't been paying close attention to reflection, so I wasn't aware
> that it was std::vector based. Maybe that ends up accelerating the need
> for someone to look into constexpr as consteval.
>
> On Mon, Feb 5, 2024 at 12:03 PM Patrice Roy via SG14 <
> sg14_at_[hidden]> wrote:
>
>> If we rely on vector but in a consteval context, I don't think there's an
>> issue (I wonder if that subtlety is discussed when examining what goes in
>> freestanding; I don't remember, personally) since it gets rid of the costs
>> for memory allocation as well as those for exception handling. It would be
>> cool if freestanding distinguished runtime and consteval contexts for these
>> tools. Unless there are costs I'm missing here (that's possible)...
>>
>> Le ven. 2 févr. 2024 à 18:58, Paul M. Bendixen via SG14 <
>> sg14_at_[hidden]> a écrit :
>>
>>> Hello all
>>>
>>> Just wanted to mention something that might be interesting for the
>>> February meeting.
>>>
>>> I recently saw a comment on the current state of reflection that seems
>>> to be moving again ( https://wg21.link/p2996 ).
>>> The worry was that the interface seems to rely on std::vector, and where
>>> it relates to strings, it uses std::string_view,
>>> maybe a std::span or even std::array could be used instead?
>>> Is this (and other parts of the reflection proposal) something we should
>>> consider during the feb meeting?
>>>
>>> Michael also mentioned it might be interesting to get Ben Craig on to
>>> talk about what's going on in freestanding?
>>>
>>> Best regards
>>> Paul M. Bendixen
>>>
>>> • − − •/• −/• • −/• − • •/− • • •/•/− •/− • •/• •/− • • −/•/− •/• − − •−
>>> •/− − •/− −/• −/• •/• − • •/• − • − • −/− • − •/− − −/− −//
>>> _______________________________________________
>>> SG14 mailing list
>>> SG14_at_[hidden]
>>> https://lists.isocpp.org/mailman/listinfo.cgi/sg14
>>>
>> _______________________________________________
>> SG14 mailing list
>> SG14_at_[hidden]
>> https://lists.isocpp.org/mailman/listinfo.cgi/sg14
>>
> _______________________________________________
> SG14 mailing list
> SG14_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/sg14
>
Embedded. I would like to add all these to the agenda. Cheers.
On Mon, Feb 5, 2024 at 4:46 PM Ben Craig via SG14 <sg14_at_[hidden]>
wrote:
> Treating constexpr functions and containers as consteval on freestanding
> is on my roadmap, but I haven't started doing any of the prototyping or
> testing necessary to write a paper.
>
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2020/p2268r0.html#constexpr_to_consteval
>
> I haven't been paying close attention to reflection, so I wasn't aware
> that it was std::vector based. Maybe that ends up accelerating the need
> for someone to look into constexpr as consteval.
>
> On Mon, Feb 5, 2024 at 12:03 PM Patrice Roy via SG14 <
> sg14_at_[hidden]> wrote:
>
>> If we rely on vector but in a consteval context, I don't think there's an
>> issue (I wonder if that subtlety is discussed when examining what goes in
>> freestanding; I don't remember, personally) since it gets rid of the costs
>> for memory allocation as well as those for exception handling. It would be
>> cool if freestanding distinguished runtime and consteval contexts for these
>> tools. Unless there are costs I'm missing here (that's possible)...
>>
>> Le ven. 2 févr. 2024 à 18:58, Paul M. Bendixen via SG14 <
>> sg14_at_[hidden]> a écrit :
>>
>>> Hello all
>>>
>>> Just wanted to mention something that might be interesting for the
>>> February meeting.
>>>
>>> I recently saw a comment on the current state of reflection that seems
>>> to be moving again ( https://wg21.link/p2996 ).
>>> The worry was that the interface seems to rely on std::vector, and where
>>> it relates to strings, it uses std::string_view,
>>> maybe a std::span or even std::array could be used instead?
>>> Is this (and other parts of the reflection proposal) something we should
>>> consider during the feb meeting?
>>>
>>> Michael also mentioned it might be interesting to get Ben Craig on to
>>> talk about what's going on in freestanding?
>>>
>>> Best regards
>>> Paul M. Bendixen
>>>
>>> • − − •/• −/• • −/• − • •/− • • •/•/− •/− • •/• •/− • • −/•/− •/• − − •−
>>> •/− − •/− −/• −/• •/• − • •/• − • − • −/− • − •/− − −/− −//
>>> _______________________________________________
>>> SG14 mailing list
>>> SG14_at_[hidden]
>>> https://lists.isocpp.org/mailman/listinfo.cgi/sg14
>>>
>> _______________________________________________
>> SG14 mailing list
>> SG14_at_[hidden]
>> https://lists.isocpp.org/mailman/listinfo.cgi/sg14
>>
> _______________________________________________
> SG14 mailing list
> SG14_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/sg14
>
Received on 2024-02-07 16:25:35