A unified semantics for future Erlang

Abstract

The formal semantics of Erlang is a bit too complicated to be easily understandable. Much of this complication stems from the desire to accurately model the current implementations (Erlang/OTP R11-R14), which include features (and optimizations) developed during more than two decades. The result is a two-tier semantics where systems, and in particular messages, behave differently in a local and a distributed setting. With the introduction of multi-core hardware, multiple run-queues and efficient SMP support, the boundary between local and distributed is diffuse and should ultimately be removed. In this paper we develop a new, much cleaner semantics, for such future implementations of Erlang. We hope that this paper can stimulate some much needed debate regarding a number of poorly understood features of current and future implementations of Erlang.

Towards Symbolic Execution in Erlang

“…Some of the first attempts were done by Huch [9] and, more extensively, by Fredlund [6]. More recent approaches focus on the definition of the distributed aspects of the Erlang semantics, like [4]; this semantics was later refined in [15] and [14], where some assumptions on the future of the language…”

“…Unfortunately, there is no commonly accepted semantics and, moreover, most of the above papers only cover part of the language semantics (e.g., [4,15,14] are mainly oriented towards the concurrent features of the language). Therefore, we have recently introduced a semantics for a subset of Erlang in [16].…”

“…Therefore, we have recently introduced a semantics for a subset of Erlang in [16]. In the following, we present a more elegant and general version of this semantics that follows some of the ideas in [14].…”

“…An Erlang system is a pair (Π, Q), where Π is a pool of processes and Q is the system mailbox (analogous to the ether in the semantics of [14]). We assume no order in Π since it is not relevant to our purposes (i.e., we will be interested in exploring all possible schedulings within symbolic execution).…”

“…The system mailbox Q is a set of triples (p, p , q), where q is a list of messages (values) sent from the process with pid p to the process with pid p . The system mailbox is needed to correctly model a multi-node distributed system (see the discussion in [14]). Basically, Erlang only requires that the messages sent directly between two processes must arrive in the same order.…”