In this paper we argue that pointer-induced aliasing is largely a self inflicted wound, caused by the almost universal practice in programming to transfer information by copy. As a remedy for this defect we intro duce a concept unshareable objects, and the companion concept of unique pointers, which can be moved from one place to another, but which cannot be copied. We argue that unshareable objects can be employed conveniently in many, if not most, situations where dynamic objects are being used, and without incurring their pitfalls. And we show that it takes no more than minor, and virtually costless, modifications to a typical imperative programming language to support such objects.

Another natural application of u-objects [unshareable objects — прим. моё] is discussed in [8], where we show how such objects can be used to implement tokens — objects that, like the capabilities of operating systems, represent certain authority. Such unshareable tokens can be utilized, in particular, for the control of sharing in software systems such as object-oriented databases.

Why would Rust provide fetch_max as a built-in intrinsic? Intrinsics usually exist to leverage specific hardware instructions. But x86-64 doesn't have an atomic max instruction. So there had to be a CAS loop somewhere in the pipeline.

<...>

So I started digging. What I found was a fascinating journey through five distinct layers of compiler transformations, each one peeling back another level of abstraction, until I found exactly where that loop materialized. Let me share what I discovered.

The division instruction is one of the most expensive instruction in your CPU. Thus optimizing compilers often compile divisions by known constants down to a multiplication followed by a shift. However, in some lucky cases, the compiler does not even need a shift. I call the corresponding divisors ideal.