[imellag]

What

ConditionDeadline in SyncOnAddress.cpp (from 6f6c18b) reads the condition's clockId through std::atomic_ref<const std::uint32_t>. atomic_ref of a const type is a C++26 addition that libc++ doesn't implement yet, so the file stops compiling on macOS (atomic_ref.h:151: cannot initialize a parameter of type 'unsigned int *' with an lvalue of type 'const unsigned int *'). libstdc++ accepts it, which is why Linux and MinGW build.

Its only caller, ConditionWait, already passes a mutable UserCondition*. So the parameter drops const, and the read uses std::atomic_ref<std::uint32_t> like the other fields in the file. The load is the same.

Tested

macOS 27.2, Apple M4 Pro, Apple clang 21, x86-64 Release (Rosetta 2). main doesn't build the libraries on macOS by itself, so the tree is main b28bb77217 + the open macOS PRs #11554, #7631, #7804, #4630, #14883 and #15002, plus this commit:

• SyncOnAddress.cpp and the full cmake --build / --target libs compile;
• ctest -R 'umtx|guest_cond_timedwait|guest_ult_condition|guest_ult_sync_rwlock': 14 of 14 pass, including guest_umtx_op, whose absolute CLOCK_MONOTONIC waits go through this read.

Not run on Linux or Windows; the change there is the same load through a non-const reference.

Checklist

• Based on current main; no other open PR implements the same functions
• One topic per PR; follow-ups go in a new PR
• No comments in code except technical debt
• Unimplemented paths throw (NotImplemented_nid_no_patch); silent stubs are listed under Silent stubs in their library's technical debt file
• No notes, investigation .md files or images added to the repository (attach them to this PR)
• New third-party code is a submodule built from source
• Generic behaviour, not specific to one title
• Depends on: none
• Fixes: none
• Unblocks: none
• GPUs tested: n/a
• AI-assisted: yes