[ruby-core:119281] [Ruby master Bug#20758] Build failure with macOS Sequoia (15.0) and Xcode 16

[ruby-core:119281] [Ruby master Bug#20758] Build failure with macOS Sequoia (15.0) and Xcode 16

[eightbitraptor (Matthew Valentine-House) via ruby-core]

Issue #20758 has been reported by eightbitraptor (Matthew Valentine-House).

----------------------------------------
Bug #20758: Build failure with macOS Sequoia (15.0) and Xcode 16
https://bugs.ruby-lang.org/issues/20758

* Author: eightbitraptor (Matthew Valentine-House)
* Status: Open
* Backport: 3.1: UNKNOWN, 3.2: UNKNOWN, 3.3: UNKNOWN
----------------------------------------
## Observed behaviour

Test failures in `TestProcess#test_daemon_noclose`. Output as follows:

```
Run options:
  --seed=56670
  "--ruby=./miniruby -I../src/lib -I. -I.ext/common  ../src/tool/runruby.rb --extout=.ext  -- --disable-gems"
  --excludes-dir=../src/test/.excludes
  --name=!/memory_leak/

# Running tests:

[ 99/158] TestProcess#test_daemon_nocloseobjc[89134]: +[NSPlaceholderMutableString initialize] may have been in progress in another thread when fork() was called.
objc[89134]: +[NSPlaceholderMutableString initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork() child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug.
 = 0.00 s
  1) Failure:
TestProcess#test_daemon_noclose [/Users/mattvh/git/ruby/src/test/ruby/test_process.rb:1882]:
<"ok\n" + "/\n"> expected but was
<"">.

Finished tests in 8.625520s, 18.3177 tests/s, 112.5729 assertions/s.
158 tests, 971 assertions, 1 failures, 0 errors, 1 skips

ruby -v: ruby 3.4.0dev (2024-09-23T12:23:58Z master 4c3727c6ac) +PRISM [arm64-darwin24]
make: *** [yes-test-all] Error 1

```

## Expected behaviour

I expect the tests to all pass with no errors.

## Notes

This started happening for me immediately after upgrading my laptop to the latest macOS using the software update tool. Versions reported are: 

```
Command Line Tools for Xcode                       16.0       23/09/2024, 14:24:42
macOS Sequoia 15.0                                 15.0       23/09/2024, 14:41:14
```

This appears to be the same issue as seen in https://bugs.ruby-lang.org/issues/18912

I can replicate it using the same minimal repro that @katei described in https://bugs.ruby-lang.org/issues/18912#note-15

```
Process.daemon(false, true)
Dir.pwd
```

and can confirm that exporting the variable `OBJC_DISABLE_INITIALIZE_FORK_SAFETY=YES` suppresses the error.

This behaviour was tested on this commit https://github.com/ruby/ruby/commit/4c3727c6ac



-- 
https://bugs.ruby-lang.org/
 ______________________________________________
 ruby-core mailing list -- ruby-core@ml.ruby-lang.org
 To unsubscribe send an email to ruby-core-leave@ml.ruby-lang.org
 ruby-core info -- https://ml.ruby-lang.org/mailman3/lists/ruby-core.ml.ruby-lang.org/

[ruby-core:119282] [Ruby master Bug#20758] Build failure with macOS Sequoia (15.0) and Xcode 16

[eightbitraptor (Matthew Valentine-House) via ruby-core]

Issue #20758 has been updated by eightbitraptor (Matthew Valentine-House).

Status changed from Open to Closed

Closing this as a duplicate. My apologies.

----------------------------------------
Bug #20758: Build failure with macOS Sequoia (15.0) and Xcode 16
https://bugs.ruby-lang.org/issues/20758#change-109891

* Author: eightbitraptor (Matthew Valentine-House)
* Status: Closed
* Backport: 3.1: UNKNOWN, 3.2: UNKNOWN, 3.3: UNKNOWN
----------------------------------------
## Observed behaviour

Test failures in `TestProcess#test_daemon_noclose`. Output as follows:

```
Run options:
  --seed=56670
  "--ruby=./miniruby -I../src/lib -I. -I.ext/common  ../src/tool/runruby.rb --extout=.ext  -- --disable-gems"
  --excludes-dir=../src/test/.excludes
  --name=!/memory_leak/

# Running tests:

[ 99/158] TestProcess#test_daemon_nocloseobjc[89134]: +[NSPlaceholderMutableString initialize] may have been in progress in another thread when fork() was called.
objc[89134]: +[NSPlaceholderMutableString initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork() child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug.
 = 0.00 s
  1) Failure:
TestProcess#test_daemon_noclose [/Users/mattvh/git/ruby/src/test/ruby/test_process.rb:1882]:
<"ok\n" + "/\n"> expected but was
<"">.

Finished tests in 8.625520s, 18.3177 tests/s, 112.5729 assertions/s.
158 tests, 971 assertions, 1 failures, 0 errors, 1 skips

ruby -v: ruby 3.4.0dev (2024-09-23T12:23:58Z master 4c3727c6ac) +PRISM [arm64-darwin24]
make: *** [yes-test-all] Error 1

```

## Expected behaviour

I expect the tests to all pass with no errors.

## Notes

This started happening for me immediately after upgrading my laptop to the latest macOS using the software update tool. Versions reported are: 

```
Command Line Tools for Xcode                       16.0       23/09/2024, 14:24:42
macOS Sequoia 15.0                                 15.0       23/09/2024, 14:41:14
```

This appears to be the same issue as seen in https://bugs.ruby-lang.org/issues/18912

I can replicate it using the same minimal repro that @katei described in https://bugs.ruby-lang.org/issues/18912#note-15

```
Process.daemon(false, true)
Dir.pwd
```

and can confirm that exporting the variable `OBJC_DISABLE_INITIALIZE_FORK_SAFETY=YES` suppresses the error.

This behaviour was tested on this commit https://github.com/ruby/ruby/commit/4c3727c6ac



-- 
https://bugs.ruby-lang.org/
 ______________________________________________
 ruby-core mailing list -- ruby-core@ml.ruby-lang.org
 To unsubscribe send an email to ruby-core-leave@ml.ruby-lang.org
 ruby-core info -- https://ml.ruby-lang.org/mailman3/lists/ruby-core.ml.ruby-lang.org/