search

cplusplus / papers

ISO/IEC JTC1 SC22 WG21 paper scheduling and management
643 stars 18 forks source link

P3032 R2 Less transient constexpr allocation #1751

Open wg21bot opened 8 months ago

wg21bot commented 8 months ago

P3032R0 Less transient constexpr allocation (Barry Revzin)

hanickadot commented 7 months ago

EWG discussed P3032 in Tokyo Thursday PM.

P3032 Less transient constexpr allocation: Forward as is to CWG

| SF | F | N | A | SA | | 9 | 13 | 1 | 0 | 0 |

Consensus

jensmaurer commented 7 months ago

CWG 2024-03-22: Approved for Saturday's plenary straw polls.

jfbastien commented 7 months ago

Revisited by EWG in Tokyo on Friday

P3032R1 Less transient constexpr allocation: Forward to CWG as DR for C++20.

SF F N A SA 6 5 3 0 0

Consensus.

jensmaurer commented 6 months ago

Plenary 2024-03-23: The paper is missing a change to "permissible result of constant expression" to make the f5 example valid.

wg21bot commented 6 months ago

P3032R1 Less transient constexpr allocation (Barry Revzin)

wg21bot commented 6 months ago

P3032R2 Less transient constexpr allocation (Barry Revzin)

jfbastien commented 4 months ago

Expression case was added since EWG reviewed. Re-scheduling for EWG, will send to CWG after.

jfbastien commented 3 months ago

Seen in EWG in St Louis:

Poll: P3032R2 — Less transient constexpr allocation, move the paper as written to CWG for inclusion in C++26, which allows local std::unique_ptr, forbid mutating.

SF F N A SA
2 14 1 2 2

Poll: P3032R2 — Less transient constexpr allocation, change the paper to only allow std::vector and std::basic_string, even for persistent allocation.

SF F N A SA
4 9 8 3 0

Result: the first poll has better consensus, but it could be further increased by obtaining implementation experience. Barry will try to get such experience by the next meeting, at which point we will move the paper to CWG.