GH-49569: [CI][Python][C++] Add check targetting Apple clang on deciding whether to use std::bit_width or std::log2p1#49570
Conversation
…ciding whether to use std::bit_width or std::log2p1
|
@github-actions crossbow submit test-r-macos-as-cran test-ubuntu-22.04-cpp-emscripten |
|
|
|
Revision: 066bf57 Submitted crossbow builds: ursacomputing/crossbow @ actions-576a7cbdce
|
|
@github-actions crossbow submit test-r-macos-as-cran test-ubuntu-22.04-cpp-emscripten |
|
Revision: a982f19 Submitted crossbow builds: ursacomputing/crossbow @ actions-d0624cbd1c
|
|
Failing job is now successful: |
|
@github-actions crossbow submit test-r-macos-as-cran test-ubuntu-22.04-cpp-emscripten |
|
Revision: 4736f08 Submitted crossbow builds: ursacomputing/crossbow @ actions-025f27de49
|
kou
left a comment
There was a problem hiding this comment.
+1
I didn't know __apple_build_version__!
|
After merging your PR, Conbench analyzed the 2 benchmarking runs that have been run so far on merge-commit ba273ca. There were no benchmark performance regressions. 🎉 The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them. |
Rationale for this change
Clang 15.0.7 (/opt/homebrew/bin/clang++) - Homebrew LLVM fails compiling with:
This seems to be the case of clang not having
__cpp_lib_bitopsdefined butstd::log2p1having been removed.What changes are included in this PR?
Check for
__apple_build_version__instead of__clang__so non Apple Clang just usesstd::bit_widthAre these changes tested?
Via CI
Are there any user-facing changes?
No