Abstract:We describe Elevator, a unifying polymorphic foundation for metaprogramming with memory management based on adjoint modalities. In this setting, we distinguish between multiple memory regions using modes where each mode has a specific set of structural properties. This allows us not only to capture linear (i.e. garbage-free) memory regions and (ordinary) intuitionistic (i.e. garbage-collected or persistent) memory regions, but also to capture accessibility between the memory regions using a preorder between modes. This preorder gives us the power to describe monadic and comonadic programming. As a consequence, it extends the existing logical view of metaprogramming in two directions: first, it ensures that code generation can be done efficiently by controlling memory accesses; second, it allows us to provide resource guarantees about the generated code (i.e. code that is for example garbage-free). We present the static and dynamic semantics of Elevator. In particular, we prove the substructurality of variable references and type safety of the language. We also establish mode safety, which guarantees that the evaluation of a term does not access a value in an inaccessible memory.

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve This paper attempts to address the problem of generating type-safe, polymorphic, and resource-aware code in metaprogramming. Specifically, the authors propose a new foundational system **E/l.sc/e.sc/v.sc/a.sc/t.sc/o.sc/r.sc**, which is based on adjoint logic to handle memory management. The key issues the paper aims to solve are as follows: 1. **Type-Safe Metaprogramming**: - Existing metaprogramming languages often generate code that lacks type safety guarantees and may contain dangling free variables or type errors. The paper introduces modes and adjoint modalities to ensure that the generated code is type-safe. 2. **Resource and Memory Management**: - Traditional metaprogramming languages typically do not provide guarantees about resource and memory usage, limiting the compiler's ability to generate efficient code using this information. The paper provides resource and memory usage guarantees by distinguishing different memory regions (e.g., linear memory regions and ordinary intuitionistic memory regions) and using a partial order relationship between modes to describe access relationships between these regions. 3. **Polymorphism**: - Generating polymorphic code remains a challenge. The paper combines System-F style polymorphism with adjoint modalities to provide a core language that supports polymorphic code generation. 4. **Fine-Grained Memory Control**: - The paper introduces contextual adjoint modalities to achieve fine-grained memory control, ensuring that the generated code is efficient in memory access and avoids unnecessary memory access overhead. 5. **Operational Semantics**: - The paper defines a small-step operational semantics suitable for metaprogramming, ensuring that terms do not access unreachable memory content during evaluation, thereby improving the system's reliability and safety. ### Main Contributions 1. **Contextual Adjoint Logic**: - By introducing modes to describe different memory regions and using a partial order relationship between modes to describe access relationships between these regions, the memory access relationships are both easy to check and capable of expressing a wide range of memory regions and their relationships. 2. **Operational Semantics for Metaprogramming**: - Defines a small-step operational semantics suitable for metaprogramming, ensuring that terms do not access unreachable memory content during evaluation, thereby improving the system's reliability and safety. 3. **Adjoint Modalities and System-F Style Polymorphism**: - Combines System-F style polymorphism with adjoint modalities to provide a core language that supports polymorphic code generation and ensures that the generated code is type-safe. ### Example Application The paper demonstrates how to use the mode structure to update arrays and generate polymorphic and garbage-free code through a concrete example—In-Place Array Update. This example not only shows the theoretical feasibility but also validates it through a prototype implementation. Overall, this paper provides a new theoretical foundation for metaprogramming, extending System-F and combining it with adjoint modalities to generate type-safe, polymorphic, and resource-aware code.

Polymorphic Metaprogramming with Memory Management -- An Adjoint Analysis of Metaprogramming

Moebius: Metaprogramming using Contextual Types -- The stage where System F can pattern match on itself (Long Version)

Floorplan: Spatial Layout in Memory Management Systems

A denotational account of C11-style memory

Truly Visual Polymorphic Algebraic Data Structures through Maramafication

Program Adverbs and Tlön Embeddings

Unifying Pointer Analyses for Polyglot Inter-operations through Summary Specialization

Safely Abstracting Memory Layouts

Modal Abstractions for Virtualizing Memory Addresses

Reference Capabilities for Flexible Memory Management: Extended Version

Profile-guided meta-programming

New Lace and Arsenic: adventures in weak memory with a program logic

Effects Without Monads: Non-determinism -- Back to the Meta Language

Latent Effects for Reusable Language Components: Extended Version

Memory Consistency and Program Transformations

A Formal C Memory Model for Separation Logic

Polymorphic Type Inference for Dynamic Languages

Resource Polymorphism

Generating Code with Polymorphic let: A Ballad of Value Restriction, Copying and Sharing

Separation of concerning things: a simpler basis for defining and programming with the C/C++ memory model (extended version)

A Two-Phase Infinite/Finite Low-Level Memory Model