GH-50631: [C++] Scope macOS SDK 11.3 simdjson workaround to bundled simdjson

[Reranko05]

Rationale for this change

The workaround introduced for the macOS 11.3 SDK currently uses a global add_compile_definitions(SIMDJSON_CONCEPT_DISABLED=1).

Since this workaround is only needed when building the bundled simdjson dependency, scope it to the simdjson target inside build_simdjson() instead of applying it globally.

What changes are included in this PR?

• Move the macOS SDK 11.3 workaround into build_simdjson().
• Replace the global add_compile_definitions() with target_compile_definitions(simdjson PUBLIC SIMDJSON_CONCEPT_DISABLED=1).

• GitHub Issue: [C++] Scope macOS SDK 11.3 simdjson workaround to build_simdjson #50631

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[github-actions]

⚠️ GitHub issue #50631has been automatically assigned in GitHub to PR creator.

[Reranko05]

@kou As suggested, I opened an upstream issue for this: simdjson/simdjson#2787

[kou]

@github-actions crossbow submit -g r

[github-actions]

Revision: fa60e5d

Submitted crossbow builds: ursacomputing/crossbow @ actions-343cdd513e

Task	Status
r-binary-packages
r-recheck-most
test-r-alpine-linux-cran
test-r-arrow-backwards-compatibility
test-r-depsource-system
test-r-dev-duckdb
test-r-devdocs
test-r-extra-packages
test-r-fedora-clang
test-r-gcc-11
test-r-gcc-12
test-r-install-local
test-r-install-local-minsizerel
test-r-linux-as-cran
test-r-linux-rchk
test-r-linux-sanitizers
test-r-linux-valgrind
test-r-m1-san
test-r-macos-as-cran
test-r-offline-maximal
test-r-ubuntu-22.04
test-r-versions
test-r-wasm

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[rok]

@github-actions crossbow submit test-r-macos-as-cran

[github-actions]

Revision: 4c4d64c

Submitted crossbow builds: ursacomputing/crossbow @ actions-ec06c62a76

Task	Status
test-r-macos-as-cran

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 1 comment.

[Reranko05]

@kou@rok The original simdjson/concepts.h failure is gone after changing the definition to PUBLIC. The Crossbow job now fails later in vector_sort.cc with a different error (std::span<uint64_t>), which seems unrelated to this change. Any thoughts on this new failure?

[rok]

@Reranko05 the issue was likely caused by introduction of std::span in #50248, I opened an issue #50636. Otherwise I think the change here is good, but I defer to @kou :)

[Reranko05]

@rok Thanks for the review :)

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated no new comments.

[kou]

+1

[Reranko05]

@kou I reported this upstream as suggested. They closed the issue, since the macOS 11.3 SDK is considered an unsupported configuration in simdjson, so they don't plan to handle it there. We'll likely need to keep the workaround on the Arrow side.

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit cc90e40.

There was 1 benchmark result indicating a performance regression:

• Commit Run on amd64-m5-4xlarge-linux at 2026-07-25 15:25:50Z
  • dataframe-to-table (R) with dataset=type_dict, language=R

The full Conbench report has more details. It also includes information about 9 possible false positives for unstable benchmarks that are known to sometimes produce them.