[libcu++] Implement P4206R0 Revert string support in std::constant_wrapper

[davebayer]

This paper simplifies the implementation a lot, potentially making it useful even in C++17

[coderabbitai]

Note

Reviews paused

It looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the reviews.auto_review.auto_pause_after_reviewed_commits setting.

Use the following commands to manage reviews:

• @coderabbitai resume to resume automatic reviews.
• @coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

• ▶️ Resume reviews
• 🔍 Trigger review

📝 Walkthrough

Walkthrough

__constant_wrapper now uses an auto NTTP-based value model with rewritten operator plumbing, the feature-test comment is updated, and const.wrap.class tests adjust compiler gating plus expected conversion, indexing, type, and kernel-parameter behavior.

Changes

constant_wrapper implementation and tests

Layer / File(s)	Summary
Wrapper representation and operators libcudacxx/include/cuda/std/__utility/constant_wrapper.h	Adds auto_cast support, changes __constant_wrapper and __cw to use auto NTTPs, rewrites constexpr-parameter detection, and updates operator, callable, subscript, assignment, conversion, and pseudo-mutator result typing to use _LIBCUDACXX_AUTO_CAST.
Feature-test version text libcudacxx/include/cuda/std/version	Updates the commented __cccl_lib_constant_wrapper version literal.
Compiler gating updates libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/adl.compile.pass.cpp, assign.pass.cpp, binary_ops.pass.cpp, call.pass.cpp, comma.pass.cpp, comp.pass.cpp, convert.pass.cpp, cw.pass.cpp, general.pass.cpp, mem_ptr.pass.cpp, pseudo_mutators.pass.cpp, subscript.pass.cpp, types.compile.pass.cpp, unary_ops.pass.cpp	Removes GCC 10/11 unsupported notes, changes mem_ptr.pass.cpp to nvcc-12.0, and keeps the remaining nvcc and nvrtc exclusions unchanged.
Test expectation changes libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/convert.pass.cpp, cw.pass.cpp, subscript.pass.cpp, types.compile.pass.cpp, unary_ops.pass.cpp, kernel_parameter.array.pass.cpp, kernel_parameter.pass.cpp, cw_fixed.array.ctor.pass.cpp, cw_fixed.ctor.pass.cpp, ctad.compile.pass.cpp	Updates conversion, array, object, subscript, kernel parameter, and type assertions for the new __constant_wrapper value model, and removes obsolete fixed-value, CTAD, and string-literal coverage.

Related PRs: None specified
Suggested labels: libcudacxx, needs-review
Suggested reviewers: miscco, jrhemstad

---

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/convert.pass.cpp (1)

43-46: 🎯 Functional Correctness | 🔵 Trivial | 💤 Low value

suggestion: identity check dropped for scalar conversion. The struct case below (Line 55) still asserts &result == &cws.value to confirm the reference returned by operator const _Tp&() aliases the static storage. The scalar case now only checks the copied value, losing that coverage for the by-value/by-reference contract on scalars. Not blocking since int conversion via copy is still exercised.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 63919e06-2744-4992-b9dc-4e94cc7d4d85

📥 Commits

Reviewing files that changed from the base of the PR and between 5d12ea0 and d361c13.

📒 Files selected for processing (19)

• libcudacxx/include/cuda/std/__utility/constant_wrapper.h
• libcudacxx/include/cuda/std/version
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/adl.compile.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/assign.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/binary_ops.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/call.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/comma.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/comp.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/convert.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/ctad.compile.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/cw.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/cw_fixed.array.ctor.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/cw_fixed.ctor.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/general.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/mem_ptr.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/pseudo_mutators.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/subscript.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/types.compile.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/unary_ops.pass.cpp

💤 Files with no reviewable changes (14)

• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/assign.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/unary_ops.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/cw_fixed.array.ctor.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/cw_fixed.ctor.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/general.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/ctad.compile.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/binary_ops.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/comp.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/comma.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/adl.compile.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/mem_ptr.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/call.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/subscript.pass.cpp
• libcudacxx/test/libcudacxx/std/utilities/const.wrap.class/pseudo_mutators.pass.cpp

[github-actions]

⏱️ CCCL compile-time benchmark comparison: Public headers compile-time bench

Result: 4 regression row(s), 1 improvement row(s) above threshold.

Run	Value
Config	public-headers-gcc13
Baseline	origin/main
Preset	all-dev
Targets	cub.headers.base, thrust.cpp.cuda.headers.base, libcudacxx.test.public_headers
GPU / launch args	rtx2080 / --cuda 13.3 --host gcc13

Artifacts: reports and traces

Direct file processing

-f file-processing exclusive --sort total

🔴 Direct file processing — Regressions

Rank	Regression impact	Selected Δ	Baseline	Current	Event	Matched traces
1	1.440989	1.440989	19.433520	20.874509	Processing Header File: libcudacxx/include/cuda/__device/physical_device.h	94
2	0.284245	0.284245	1.900187	2.184432	Processing Header File: cub/cub/device/dispatch/tuning/tuning_scan.cuh	68
3	0.280132	0.280132	1.826128	2.106260	Processing Header File: libcudacxx/include/cuda/std/string_view	98
4	0.268395	0.268395	7.608120	7.876515	Processing Header File: cub/cub/util_type.cuh	279

🟢 Direct file processing — Improvements

Rank	Improvement impact	Selected Δ	Baseline	Current	Event	Matched traces
1	1.101654	-1.101654	9.609696	8.508042	Processing Header File: libcudacxx/include/cuda/std/__cccl/prologue.h	550

[github-actions]

🥳 CI Workflow Results

🟩 Finished in 3h 25m: Pass: 100%/115 | Total: 4d 06h | Max: 3h 25m | Hits: 56%/940169

See results here.