Submitter: David Keaton (suggested by Hans Boehm)
Submission Date: 2014-07-16
Reference Document: N/A
Date: June 2014
Subject: Fixing an inconsistency in
The C committee intended to adopt the same model for atomics as C++ to ensure compatibility. Somewhere along the way, there was an error in synchronizing with the C++ atomic model. This could have serious consequences for code that needs to share atomic objects between modules written in C and modules written in C++ (for example, in the case of libraries written in one language being used by a program written in the other).
The C++ standard states the following in 29.4p2.
atomic_is_lock_free(29.6) indicates whether the object is lock-free. In any given program execution, the result of the lock-free query shall be consistent for all pointers of the same type.
However, the C standard states the following in 22.214.171.124p3.
atomic_is_lock_freegeneric function returns nonzero (true) if and only if the object's operations are lock-free. The result of a lock-free query on one object cannot be inferred from the result of a lock-free query on another object.
The primary issue is compatibility. Secondarily, if the
lock-free property for a given pointer type can change after an
algorithm starts, then
atomic_is_lock_free cannot be used
to select an algorithm in advance if the algorithm will allocate new
atomic objects. The C++ model is therefore more useful.
The error in synchronizing with C++ should be fixed by correcting
the behavior of
atomic_is_lock_free to be the same in C
as in C++.
Suggested Technical Corrigendum
Replace the following sentence from 126.96.36.199p3
The result of a lock-free query on one object cannot be inferred from the result of a lock-free query on another object.
with the following.
In any given program execution, the result of the lock-free query shall be consistent for all pointers of the same type.