Search

Today's software systems are too complex to ensure security after the fact – security has to be built into systems by design. To this end, model-based techniques such as UMLsec support the design-time specification and analysis of security requirements by providing custom model annotations and checks. Yet, a particularly challenging type of complexity arises from the variability of software product lines. Analyzing the security of all products separately is generally infeasible. In this work, we propose SecPL, a methodology for ensuring security in a software product line. SecPL allows developers to annotate the system design model with product-line variability and security requirements. To keep the exponentially large configuration space tractable during security checks, SecPL provides a family-based security analysis. In our experiments, this analysis outperforms the naive strategy of checking all products individually. Finally, we present the results of a user study that indicates the usability of our overall methodology.

Regular expressions are part of every programmer’s toolbox. They are used for a wide variety of language-related tasks and there are many algorithms for manipulating them. In particular, matching algorithms that detect whether a word belongs to the language described by a regular expression are well explored, yet new algorithms appear frequently. However, there is no satisfactory methodology for testing such matchers. We propose a testing methodology which is based on generating positive as well as negative examples of words in the language. To this end, we present a new algorithm to generate the language described by a generalized regular expression with intersection and complement operators. The complement operator allows us to generate both positive and negative example words from a given regular expression. We implement our generator in Haskell and OCaml and show that its performance is more than adequate for testing.

Relational database management systems (RDBMS) are operationally similar to a dynamic language processor. They take SQL queries as input, dynamically generate an optimized execution plan, and then execute it. In recent decades, the emergence of in-memory databases with columnar storage, which use array-like storage structures, has shifted the focus on optimizations from the traditional I/O bottleneck to CPU and memory. However, database research so far has primarily focused on CPU cache optimizations. The similarity in the computational characteristics of such database workloads and array programming language optimizations are largely unexplored. We believe that these database implementations can benefit from merging database optimizations with dynamic array-based programming language approaches. Therefore, in this paper, we propose a novel approach to optimize database query execution using a new array-based intermediate representation, HorseIR, that resides between database queries and compiled code. Furthermore, we provide a translator to generate HorseIR from database execution plans and a compiler that optimizes HorseIR and generates efficient code. We compare HorseIR with the MonetDB RDBMS, by testing standard SQL queries, and show how our approach and compiler optimizations improve the runtime of complex queries.

Logging is a fundamental part of the software development and deployment lifecycle but logging support is often provided as an afterthought via limited library APIs or third-party modules. Given the critical nature of logging in modern cloud, mobile, and IoT development workflows, the unique needs of the APIs involved, and the opportunities for optimization using semantic knowledge, we argue logging should be included as a central part of the language and runtime designs. This paper presents a rethinking of the logger for modern cloud-native workflows. Based on a set of design principles for modern logging we build a logging system, that supports near zero-cost for disabled log statements, low cost lazy-copying for enabled log statements, selective persistence of logging output, unified control of logging output across different libraries, and DevOps integration for use with modern cloud-based deployments. To evaluate these concepts we implemented the Log++ logger for Node.js hosted JavaScript applications.

There are several different gradual typing semantics, reflecting different trade-offs between performance and type soundness guarantees. Notably absent, however, are any data on which of these semantics developers actually prefer. We begin to rectify this shortcoming by surveying professional developers, computer science students, and Mechanical Turk workers on their preferences between three gradual typing semantics. These semantics reflect important points in the design space, corresponding to the behaviors of Typed Racket, TypeScript, and Reticulated Python. Our most important finding is that our respondents prefer a runtime semantics that fully enforces statically declared types.

Recent advances in execution environments for JavaScript and WebAssembly that run on a broad range of devices, from workstations and mobile phones to IoT devices, provide new opportunities for portable and web-based numerical computing. Indeed, numerous numerical libraries and applications are emerging on the web, including Tensorflow.js, JSMapReduce, and the NLG Protein Viewer. This paper evaluates the current performance of numerical computing on the web, including both JavaScript and WebAssembly, over a wide range of devices from workstations to IoT devices. We developed a new benchmarking approach, which allowed us to perform centralized benchmarking, including benchmarking on mobile and IoT devices. Using this approach we performed four performance studies using the Ostrich benchmark suite, a collection of numerical programs representing the numerical dwarf categories identified by Colella. We studied the performance evolution of JavaScript, the relative performance of WebAssembly, the performance of server-side Node.js, and a comprehensive performance showdown for a wide range of devices.

Operating systems are traditionally implemented in low- level, performance-oriented programming languages. These languages typically rely on minimal runtime support and provide unfettered access to the underlying hardware. Tra- dition has benefits: developers control the resources that the operating system manages and few performance bottle- necks cannot be overcome with clever feats of programming. On the other hand, this makes operating systems harder to understand and maintain. Furthermore, those languages have few built-in barriers against bugs. This paper is an ex- periment in side-stepping operating systems, and pushing functionality into the runtime of high-level programming languages. The question we try to answer is how much sup- port is needed to run an application written in, say, Smalltalk or Python on bare metal, that is, with no underlying oper- ating system. We present a framework named NopSys that allows this, and we validate it with the implementation of CogNos a Smalltalk virtual machine running on bare x86 hardware. Experimental results suggest that this approach is promising.

Library authors often are faced with a design choice: should a function with preconditions be implemented as a partial function, or by returning a failure condition on incorrect use? Neither option is ideal. Partial functions lead to frustrating run-time errors. Failure conditions must be checked at the use-site, placing an unfair tax on the users who have ensured that the function's preconditions were correctly met. In this paper, we introduce an API design concept called ``ghosts of departed proofs'' based on the following observation: sophisticated preconditions can be encoded in Haskell's type system with no run-time overhead, by using proofs that inhabit phantom type parameters attached to newtype wrappers. The user expresses correctness arguments by constructing proofs to inhabit these phantom types. Critically, this technique allows the library user to decide when and how to validate that the API's preconditions are met. The ``ghosts of departed proofs'' approach to API design can achieve many of the benefits of dependent types and refinement types, yet only requires some minor and well-understood extensions to Haskell 2010. We demonstrate the utility of this approach through a series of case studies, showing how to enforce novel invariants for lists, maps, graphs, shared memory regions, and more.

Processing data at different rates is generally a hard problem in reactive programming. Buffering problems, lags, and concurrency issues often occur. Many of these problems are clock errors , where data at different rates is combined incorrectly. Techniques to avoid clock errors, such as type-level clocks and deterministic scheduling, exist in the field of synchronous programming, but are not implemented in general-purpose languages like Haskell. Rhine is a clock-safe library for synchronous and asynchronous Functional Reactive Programming (FRP). It separates the aspects of clocking, scheduling and resampling from each other, and ensures clock-safety at the type level. Concurrent communication is encapsulated safely. Diverse reactive subsystems can be combined in a coherent, declarative data-flow framework, while correct interoperability of data at different rates is guaranteed by type-level clocks. This provides a general-purpose framework that simplifies multi-rate FRP systems and can be used for game development, media applications, GUIs and embedded systems, through a flexible API with many reusable components.

Datatype-generic programming is a widely used technique to define functions that work regularly over a class of datatypes. Examples include deriving serialization of data, equality or even functoriality. The state-of-the-art of generic programming still lacks handling GADTs, multiple type variables, and some other features. This paper exploits modern GHC extensions, including TypeInType, to handle arbitrary number of type variables, constraints, and existentials. We also provide an Agda model of our construction that does not require Russel’s paradox, proving the construction is consistent.

We present a high performance multicore I/O manager based on libuv for Glasgow Haskell Compiler (GHC). The new I/O manager is packaged as an ordinary Haskell package rather than baked into GHC's runtime system(GHC RTS), yet takes advantage of GHC RTS's comprehensive concurrent support, such as lightweight threads and safe/unsafe FFI options. The new I/O manager's performance is comparable with existing implementation, with greater stability under high load. It also can be easily extended to support all of libuv's callback-based APIs, allowing us to write a complete high performance I/O toolkit without spending time on dealing with OS differences or low-level I/O system calls.

Standby efficiency for connected devices is one of the priorities of the G20’s Energy Efficiency Action Plan. We propose transparent programming language mechanisms to enforce that applications remain in the deepest standby modes for the longest periods of time. We extend the programming language Céu with support for interrupt service routines and with a simple power management runtime. Based on these primitives, we also provide device drivers that allow applications to take advantage of standby automatically. Our approach relies on the synchronous semantics of the language which guarantees that reactions to the environment always reach an idle state amenable to standby. In addition, in order to lower the programming barrier of adoption, we show that programs in Céu can keep a sequential syntactic structure, even when applications require non-trivial concurrent behavior.

Systems neuroscience studies involving in-vivo models often require realtime data processing. In these studies, many events must be monitored and processed quickly, including behavior of the subject (e.g., movement of a limb) or features of neural data (e.g., a neuron transmitting an action potential). Unfortunately, most realtime platforms are proprietary, require specific architectures, or are limited to low-level programming languages. Here we present a hardware-independent, open-source realtime computation platform that supports high-level programming. The resulting platform, LiCoRICE, can process on order 10e10 bits/sec of network data at 1 ms ticks with 18.2 µs jitter. It connects to various inputs and outputs (e.g., DIO, Ethernet, database logging, and analog line in/out) and minimizes reliance on custom device drivers by leveraging peripheral support via the Linux kernel. Its modular architecture supports model-based design for rapid prototyping with C and Python/Cython and can perform numerical operations via BLAS/LAPACK-optimized NumPy that is statically compiled via Numba’s pycc. LiCoRICE is not only suitable for systems neuroscience research, but also for applications requiring closed-loop realtime data processing from robotics and control systems to interactive applications and quantitative financial trading.

JavaScript has seen meteoric growth in popularity as it has in- creasingly become the language of choice for developers, both for front-end web development and server code development through various JavaScript frameworks and Node.js. Part of the reason for its wide use is that it is a prototype based language with dynamic types, making it easy to learn and program in. This flexibility and ease of programming comes at the cost of performance. There are two sources of significant slowdown. First, since the number and type of properties of prototypes is dynamic, accessing a property involves a slow dictionary lookup, as opposed to it being present at a fixed offset from the base address. Second, the dynamism in type of values necessitates wrapping and unwrapping of values into objects with a variety of checks including for type of the value. To mitigate these performance problems, this paper proposes JSCore, a core specialized for JavaScript execution, that vastly reduces the performance degradation due to the above two causes. It uses a hardware lookup table to accelerate property access, and extends the data path to store data types with the data, nearly eliminating the second source of slowdown. Combining the two, JSCore accelerates real world JavaScript applications by 23%.

Despite the presence of garbage collection in managed languages like JavaScript, memory leaks remain a serious problem. In the context of web applications, these leaks are especially pervasive and difficult to debug. Web application memory leaks can take many forms, including failing to dispose of unneeded event listeners, repeatedly injecting iframes and CSS files, and failing to call cleanup routines in third-party libraries. Leaks degrade responsiveness by increasing GC frequency and overhead, and can even lead to browser tab crashes by exhausting available memory. Because previous leak detection approaches designed for conventional C, C++ or Java applications are ineffective in the browser environment, tracking down leaks currently requires intensive manual effort by web developers. This paper introduces BLeak (Browser Leak debugger), the first system for automatically debugging memory leaks in web applications. BLeak's algorithms leverage the observation that in modern web applications, users often repeatedly return to the same (approximate) visual state (e.g., the inbox view in Gmail). Sustained growth between round trips is a strong indicator of a memory leak. To use BLeak, a developer writes a short script (17-73 LOC on our benchmarks) to drive a web application in round trips to the same visual state. BLeak then automatically generates a list of leaks found along with their root causes, ranked by return on investment. Guided by BLeak, we identify and fix over 50 memory leaks in popular libraries and apps including Airbnb, AngularJS, Google Analytics, Google Maps SDK, and jQuery. BLeak's median precision is 100%; fixing the leaks it identifies reduces heap growth by an average of 94%, saving from 0.5 MB to 8 MB per round trip. We believe BLeak's approach to be broadly applicable beyond web applications, including to GUI applications on desktop and mobile platforms.

Dynamic program analysis can predict data races knowable from an observed execution, but existing predictive analyses either miss races or cannot analyze full program executions. This paper presents Vindicator, a novel, sound (no false races) predictive approach that finds more data races than existing predictive approaches. Vindicator achieves high coverage by using a new, efficient analysis that finds all possible predictable races but may detect false races. Vindicator ensures soundness using a novel algorithm that checks each potential race to determine whether it is a true predictable race. An evaluation using large Java programs shows that Vindicator finds hard-to-detect predictable races that existing sound predictive analyses miss, at a comparable performance cost.

Going beyond a certain number of cores in modern architectures requires an on-chip network more scalable than conventional buses. However, employing an on-chip network in a manycore system (to improve scalability) makes the latencies of the data accesses issued by a core non-uniform. This non-uniformity can play a significant role in shaping the overall application performance. This work presents a novel compiler strategy which involves exposing architecture information to the compiler to enable an optimized computation-to-core mapping. Specifically, we propose a compiler-guided scheme that takes into account the relative positions of (and distances between) cores, last-level caches (LLCs) and memory controllers (MCs) in a manycore system, and generates a mapping of computations to cores with the goal of minimizing the on-chip network traffic. The experimental data collected using a set of 21 multi-threaded applications reveal that, on an average, our approach reduces the on-chip network latency in a 6×6 manycore system by 38.4% in the case of private LLCs, and 43.8% in the case of shared LLCs. These improvements translate to the corresponding execution time improvements of 10.9% and 12.7% for the private LLC and shared LLC based systems, respectively.

On-stack replacement (OSR) is a performance-critical technology for many languages, especially dynamic languages. Conventional wisdom, apparent in JavaScript engines such as V8 and SpiderMonkey, is that OSR must be implemented in a low-level (i.e., in assembly) and language-specific way. This paper presents an OSR abstraction based on Swapstack, materialized as the API for a low-level virtual machine, and shows how the abstraction of resumption protocols facilitates an elegant implementation of this API on real hardware. Using an experimental JavaScript implementation, we demonstrate that this API enables the language implementation to perform OSR without the need to deal with machine-level details. We also show that the API itself is implementable on concrete hardware. This work helps crystallize OSR abstractions and, by providing a reusable implementation, brings OSR within reach for more language implementers.

Dynamic data race detectors are valuable tools for testing and validating concurrent software, but to achieve good performance they are typically implemented using sophisticated concurrent algorithms. Thus, they are ironically prone to the exact same kind of concurrency bugs they are designed to detect. To address these problems, we have developed V erified FT, a clean slate redesign of the F ast T rack race detector [19]. The V erified FT analysis provides the same precision guarantee as F ast T rack , but is simpler to implement correctly and efficiently, enabling us to mechanically verify an implementation of its core algorithm using CIVL [27]. Moreover, V erified FT provides these correctness guarantees without sacrificing any performance over current state-of-the-art (but complex and unverified) F ast T rack implementations for Java.

JavaScript code running in the Node.js runtime is a major platform for developers building cloud, mobile, or IoT applications. A fundamental concept in Node.js programming is the use of asynchronous callbacks and event loops to provide highly responsive applications. While conceptually simple, this programming model contains numerous subtleties and behaviors that are defined implicitly by the current Node.js implementation. This paper presents the first comprehensive formalization of the Node.js asynchronous execution model and defines a high-level notion of async-contexts to formalize fundamental relationships between asynchronous executions in an application. These formalizations provide a foundation for the construction of static or dynamic program analysis tools, support the exploration of alternative Node.js event loop implementations, and provide a high-level conceptual framework for reasoning about relationships between the execution of asynchronous callbacks in a Node.js application.

Staging has proved a successful technique for programmatically removing code abstractions, thereby allowing for faster program execution while retaining a high-level interface for the programmer. Unfortunately, techniques based on staging suffer from a number of problems — ranging from practicalities to fundamental limitations — which have prevented their widespread adoption. We introduce Quoted Staged Rewriting (QSR), an approach that uses type-safe, pattern matching-enabled quasiquotes to define optimizations. The approach is “staged” in two ways: first, rewrite rules can execute arbitrary code during pattern matching and code reconstruction, leveraging the power and flexibility of staging; second, library designers can orchestrate the application of successive rewriting phases (stages). The advantages of using quasiquote-based rewriting are that library designers never have to deal directly with the intermediate representation (IR), and that it allows for non-intrusive optimizations — in contrast with staging, it is not necessary to adapt the entire library and user programs to accommodate optimizations. We show how Squid, a Scala macro-based framework, enables QSR and renders library-defined optimizations more practical than ever before: library designers write domain-specific optimizers that users invoke transparently on delimited portions of their code base. As a motivating example we describe an implementation of stream fusion (a well-known deforestation technique) that is both simpler and more powerful than the state of the art, and can readily be used by Scala programmers with no knowledge of metaprogramming.

Building a reusable, auto-tuning code generator from scratch is a challenging problem, requiring many careful design choices. We describe HSpiral, a Haskell compiler for signal transforms that builds on the foundational work of Spiral. Our design leverages many Haskell language features to ensure that our framework is reusable, flexible, and efficient. As well as describing the design of our system, we show how to extend it to support new classes of transforms, including the number-theoretic transform and a variant of the split-radix algorithm that results in reduced operation counts. We also show how to incorporate rewrite rules into our system to reproduce results from previous literature on code generation for the fast Fourier transform. Although the Spiral project demonstrated significant advances in automatic code generation, it has not been widely used by other researchers. HSpiral is freely available under an MIT-style license, and we are actively working to turn it into a tool to further both our own research goals and to serve as a foundation for other research groups' work in developing new implementations of signal transform algorithms.

Metaprogramming is among the most promising candidates to solve the abstraction vs performance trade-off that plagues software engineering through specialization. Metaprogramming has been used to enable low-overhead generic programming for a long time, with C++ templates being one of the most prominent examples. But often a single, fixed pattern of specialization is not enough, and more flexibility is needed. Hence, this paper seeks to apply generic programming techniques to challenges in metaprogramming, in particular to abstract over the execution stage of individual program expressions. We thus extend the scope of generic programming into the dimension of time. The resulting notion of stage polymorphism enables novel abstractions in the design of program generators, which we develop and explore in this paper. We present one possible implementation, in Scala using the lightweight modular staging (LMS) framework, and apply it to two important case studies: convolution on images and the fast Fourier transform (FFT).

Monads and applicative functors are two ways that Haskell programmers bundle effectful primitives into effectful program fragments. In this paper, we investigate using monads and applicative functors to bundle remote effectful primitives, specifically aiming to amortize the cost of remote communications using bundling. We look at several ways of maximizing the bundling of primitives, drawing from the remote monad design pattern and Haxl system, and provide a taxonomy of mechanism for amortization, with examples. The result of this investigation is that monadic fragments can be efficiently bundled into packets, almost for free, when given a user-supplied packet transportation mechanism, and the primitives obey some simple pre- and post-conditions.

We present a novel but remarkably simple formulation of formal language grammars in Haskell as functions mapping a record of pro- duction parsers to itself. Thus formulated grammars are first-class objects, composable and reusable. We also provide a simple parser implementation for them, based on an improved packrat algorithm. In order to make the grammar manipulation code reusable, we introduce a set of type classes mirroring the existing type classes from Haskell base library, but whose methods have rank-2 types.

This paper presents Speculate, a tool that automatically conjectures laws involving conditional equations and inequalities about Haskell functions. Speculate enumerates expressions involving a given collection of Haskell functions, testing to separate those expressions into apparent equivalence classes. Expressions in the same equivalence class are used to conjecture equations. Representative expressions of different equivalence classes are used to conjecture conditional equations and inequalities. Speculate uses lightweight equational reasoning based on term rewriting to discard redundant laws and to avoid needless testing. Several applications demonstrate the effectiveness of Speculate.

We present a library in Haskell for programming Field Programmable Gate Arrays (FPGAs), including hardware software co-design. Code for software (in C) and hardware (in VHDL) is generated from a single program, along with the code to support communication between hardware and software. We present type-based techniques for the simultaneous implementation of more than one embedded domain specific language (EDSL). We build upon a generic representation of imperative programs that is loosely coupled to instruction and expression types, allowing the individual parts to be developed and improved separately. Code generation is implemented as a series of translations between progressively smaller, typed EDSLs, safeguarding against errors that arise in untyped translations. Initial case studies show promising performance.

Correctness and performance are often at odds in the field of systems engineering, either because correct programs are too costly to write or impractical to execute, or because well-performing code involves so many tricks of the trade that formal analysis is unable to isolate the main properties of the algorithm. As a prime example of this tension, Coq is an established proof environment that allows writing correct, dependently-typed code, but it has been criticized for exorbitant development times, forcing the developer to choose between optimal code or tractable proofs. On the other side of the divide, Haskell has proven itself to be a capable, well-typed programming environment, yet easy-to-read, straightforward code must all too often be replaced by highly optimized variants that obscure the author's original intention. This paper builds on the existing Fiat refinement framework to bridge this divide, demonstrating how to derive a correct-by-construction implementation that meets (or exceeds) the performance characteristics of highly optimized Haskell, starting from a high-level Coq specification. To achieve this goal, we extend Fiat with a stateful notion of refinement of abstract data types and add support for extracting stateful code via a free monad equipped with an algebra of heap-manipulating operations. As a case study, we reimplement a subset of the popular bytestring library, with little to no loss of performance, while retaining a high guarantee of program correctness.

We demonstrate for the first time that Liquid Haskell, a refinement type checker for Haskell programs, can be used for arbitrary theorem proving by verifying a parallel, monoidal string matching algorithm implemented in Haskell. We use refinement types to specify correctness properties, Haskell terms to express proofs of these properties, and Liquid Haskell to check the proofs. We evaluate Liquid Haskell as a theorem prover by replicating our 1428 LoC proof in a dependently-typed language (Coq - 1136 LoC). Finally, we compare both proofs, uncovering the relative advantages and disadvantages of the two provers.

Functional Reactive Programming (FRP) allows interactive applications to be modelled in a declarative manner using time-varying values. For practical reasons, however, operational constraints are often imposed, such as having a fixed time domain, time always flowing forward, and limiting the exploration of the past. In this paper we show how these constraints can be overcome, giving local control over the time domain, the direction of time and the sampling step. We study the behaviour of FRP expressions when time flows backwards, and demonstrate how to synchronize subsystems running asynchronously and at different sampling rates. We have verified the practicality of our approach with two non-trivial games in which time control is central to the gameplay.

Music description and generation are popular use cases for Haskell, ranging from live coding libraries to automatic harmonisation systems. Some approaches use probabilistic methods, others build on the theory of Western music composition, but there has been little work done on checking the correctness of musical pieces in terms of voice leading, harmony, and structure. Haskell's recent additions to the type-system now enable us to perform such analysis statically. We present our experience of implementing a type-level model of classical music and an accompanying EDSL which enforce the rules of classical music at compile-time, turning composition mistakes into compiler errors. Along the way, we discuss the strengths and limitations of doing this in Haskell and demonstrate that the type system of the language is fully capable of expressing non-trivial and practical logic specific to a particular domain.

Production compilers commonly perform dozens of transformations on an intermediate representation. Running those transformations in separate passes harms performance. One approach to recover performance is to combine transformations by hand in order to reduce number of passes. Such an approach harms modularity, and thus makes it hard to maintain and evolve a compiler over the long term, and makes reasoning about performance harder. This paper describes a methodology that allows a compiler writer to define multiple transformations separately, but fuse them into a single traversal of the intermediate representation when the compiler runs. This approach has been implemented in a compiler for the Scala language. Our performance evaluation indicates that this approach reduces the running time of tree transformations by 35% and shows that this is due to improved cache friendliness. At the same time, the approach improves total memory consumption by reducing the object tenuring rate by 50%. This approach enables compiler writers to write transformations that are both modular and fast at the same time.

Parametric polymorphism is one of the linchpins of modern typed programming, but it comes with a real performance penalty. We describe this penalty; offer a principled way to reason about it (kinds as calling conventions); and propose levity polymorphism. This new form of polymorphism allows abstractions over calling conventions; we detail and verify restrictions that are necessary in order to compile levity-polymorphic functions. Levity polymorphism has created new opportunities in Haskell, including the ability to generalize nearly half of the type classes in GHC's standard library.

Points-to analysis for Java benefits greatly from context sensitivity. CFL-reachability and k -limited context strings are two approaches to obtaining context sensitivity with different advantages: CFL-reachability allows local reasoning about data-value flow and thus is suitable for demand-driven analyses, whereas k -limited analyses allow object sensitivity which is a superior calling context abstraction for object-oriented languages. We combine the advantages of both approaches to obtain a context-sensitive analysis that is as precise as k -limited context strings, but is more efficient to compute. Our key insight is based on a novel abstraction of contexts adapted from CFL-reachability that represents a relation between two calling contexts as a composition of transformations over contexts. We formulate pointer analysis in an algebraic structure of context transformations, which is a set of functions over calling contexts closed under function composition. We show that the context representation of context-string-based analyses is an explicit enumeration of all input and output values of context transformations. CFL-reachability-based pointer analysis is formulated to use call-strings as contexts, but the context transformations concept can be applied to any context abstraction used in k -limited analyses, including object- and type-sensitive analysis. The result is a more efficient algorithm for computing context-sensitive results for a wide variety of context configurations.

SQL is the de facto language for manipulating relational data. Though powerful, many users find it difficult to write SQL queries due to highly expressive constructs. While using the programming-by-example paradigm to help users write SQL queries is an attractive proposition, as evidenced by online help forums such as Stack Overflow, developing techniques for synthesizing SQL queries from given input-output (I/O) examples has been difficult, due to the large space of SQL queries as a result of its rich set of operators. In this paper, we present a new scalable and efficient algorithm for synthesizing SQL queries based on I/O examples. The key innovation of our algorithm is development of a language for abstract queries, i.e., queries with uninstantiated operators, that can be used to express a large space of SQL queries efficiently. Using abstract queries to represent the search space nicely decomposes the synthesis problem into two tasks: 1) searching for abstract queries that can potentially satisfy the given I/O examples, and 2) instantiating the found abstract queries and ranking the results. We have implemented this algorithm in a new tool called Scythe and evaluated it using 193 benchmarks collected from Stack Overflow. Our evaluation shows that Scythe can efficiently solve 74% of the benchmarks, most in just a few seconds, and the queries range from simple ones involving a single selection to complex queries with 6 nested subqueires.

Mainstream points-to analysis techniques for object-oriented languages rely predominantly on the allocation-site abstraction to model heap objects. We present MAHJONG, a novel heap abstraction that is specifically developed to address the needs of an important class of type-dependent clients, such as call graph construction, devirtualization and may-fail casting. By merging equivalent automata representing type-consistent objects that are created by the allocation-site abstraction, MAHJONG enables an allocation-site-based points-to analysis to run significantly faster while achieving nearly the same precision for type-dependent clients. MAHJONG is simple conceptually, efficient, and drops easily on any allocation-site-based points-to analysis. We demonstrate its effectiveness by discussing some insights on why it is a better alternative of the allocation-site abstraction for type-dependent clients and evaluating it extensively on 12 large real-world Java programs with five context-sensitive points-to analyses and three widely used type-dependent clients. MAHJONG is expected to provide significant benefits for many program analyses where call graphs are required.

The maturation of the Web platform has given rise to sophisticated and demanding Web applications such as interactive 3D visualization, audio and video software, and games. With that, efficiency and security of code on the Web has become more important than ever. Yet JavaScript as the only built-in language of the Web is not well-equipped to meet these requirements, especially as a compilation target. Engineers from the four major browser vendors have risen to the challenge and collaboratively designed a portable low-level bytecode called WebAssembly. It offers compact representation, efficient validation and compilation, and safe low to no-overhead execution. Rather than committing to a specific programming model, WebAssembly is an abstraction over modern hardware, making it language-, hardware-, and platform-independent, with use cases beyond just the Web. WebAssembly has been designed with a formal semantics from the start. We describe the motivation, design and formal semantics of WebAssembly and provide some preliminary experience with implementations.

This paper explores how fork-join parallelism, as supported by concurrency platforms such as Cilk and OpenMP, can be embedded into a compiler's intermediate representation (IR). Mainstream compilers typically treat parallel linguistic constructs as syntactic sugar for function calls into a parallel runtime. These calls prevent the compiler from performing optimizations across parallel control constructs. Remedying this situation is generally thought to require an extensive reworking of compiler analyses and code transformations to handle parallel semantics. Tapir is a compiler IR that represents logically parallel tasks asymmetrically in the program's control flow graph. Tapir allows the compiler to optimize across parallel control constructs with only minor changes to its existing analyses and code transformations. To prototype Tapir in the LLVM compiler, for example, we added or modified about 6000 lines of LLVM's 4-million-line codebase. Tapir enables LLVM's existing compiler optimizations for serial code -- including loop-invariant-code motion, common-subexpression elimination, and tail-recursion elimination -- to work with parallel control constructs such as spawning and parallel loops. Tapir also supports parallel optimizations such as loop scheduling.

We explore a programming approach for concurrency that synchronizes all accesses to shared memory by default. Synchronization takes place by ensuring that all program code runs inside atomic sections even if the program code has external side effects. Threads are mapped to atomic sections that a programmer must explicitly split to increase concurrency. A naive implementation of this approach incurs a large amount of overhead. We show how to reduce this overhead to make the approach suitable for realistic application programs on existing hardware. We present an implementation technique based on a special-purpose software transactional memory system. To reduce the overhead, the technique exploits properties of managed, object-oriented programming languages as well as intraprocedural static analyses and uses field-level granularity locking in combination with transactional I/O to provide good scaling properties. We implemented the synchronized-by-default (SBD) approach for the Java language and evaluate its performance for six programs from the DaCapo benchmark suite. The evaluation shows that, compared to explicit synchronization, the SBD approach has an overhead between 0.4% and 102% depending on the benchmark and the number of threads, with a mean (geom.) of 23.9%.

Many popular languages have a self-interpreter, that is, an interpreter for the language written in itself. So far, work on polymorphically-typed self-interpreters has concentrated on self-recognizers that merely recover a program from its representation. A larger and until now unsolved challenge is to implement a polymorphically-typed self-evaluator that evaluates the represented program and produces a representation of the result. We present F ω μ i , the first λ-calculus that supports a polymorphically-typed self-evaluator. Our calculus extends F ω with recursive types and intensional type functions and has decidable type checking. Our key innovation is a novel implementation of type equality proofs that enables us to define a versatile representation of programs. Our results establish a new category of languages that can support polymorphically-typed self-evaluators.

A notion of dimension in intersection typed λ-calculi is presented. The dimension of a typed λ-term is given by the minimal norm of an elaboration (a proof theoretic decoration) necessary for typing the term at its type, and, intuitively, measures intersection introduction as a resource. Bounded-dimensional intersection type calculi are shown to enjoy subject reduction, since terms can be elaborated in non-increasing norm under β-reduction. We prove that a multiset interpretation (corresponding to a non-idempotent and non-linear interpretation of intersection) of dimensionality corresponds to the number of simultaneous constraints required during search for inhabitants. As a consequence, the inhabitation problem is decidable in bounded multiset dimension, and it is proven to be EXPSPACE-complete. This result is a substantial generalization of inhabitation for the rank 2-fragment, yielding a calculus with decidable inhabitation which is independent of rank. Our results give rise to a new criterion (dimensional bound) for subclasses of intersection type calculi with a decidable inhabitation problem, which is orthogonal to previously known criteria, and which should have immediate applications in synthesis. Additionally, we give examples of dimensional analysis of fragments of the intersection type system, including conservativity over simple types, rank 2-types, and normal form typings, and we provide some observations towards dimensional analysis of other systems. It is suggested (for future work) that our notion of dimension may have semantic interpretations in terms of of reduction complexity.

This paper introduces QWIRE (``choir''), a language for defining quantum circuits and an interface for manipulating them inside of an arbitrary classical host language. QWIRE is minimal---it contains only a few primitives---and sound with respect to the physical properties entailed by quantum mechanics. At the same time, QWIRE is expressive and highly modular due to its relationship with the host language, mirroring the QRAM model of computation that places a quantum computer (controlled by circuits) alongside a classical computer (controlled by the host language). We present QWIRE along with its type system and operational semantics, which we prove is safe and strongly normalizing whenever the host language is. We give circuits a denotational semantics in terms of density matrices. Throughout, we investigate examples that demonstrate the expressive power of QWIRE, including extensions to the host language that (1) expose a general analysis framework for circuits, and (2) provide dependent types.

Reynolds's Abstraction theorem forms the mathematical foundation for data abstraction. His setting was the polymorphic lambda calculus. Today, many modern languages, such as the ML family, employ rich module systems designed to give more expressive support for data abstraction than the polymorphic lambda calculus, but analogues of the Abstraction theorem for such module systems have lagged far behind. We give an account of the Abstraction theorem for a modern module calculus supporting generative and applicative functors, higher-order functors, sealing, and translucent signatures. The main issues to be overcome are: (1) the fact that modules combine both types and terms, so they must be treated as both simultaneously, (2) the effect discipline that models the distinction between transparent and opaque modules, and (3) a very rich language of type constructors supporting singleton kinds. We define logical equivalence for modules and show that it coincides with contextual equivalence. This substantiates the folk theorem that modules are good for data abstraction. All our proofs are formalized in Coq.

jQuery is the most popular JavaScript library but the state-of-the-art static analyzers for JavaScript applications fail to analyze simple programs that use jQuery. In this paper, we present a novel abstract string domain whose elements are simple regular expressions that can represent prefix, infix, and postfix substrings of a string and even their sets. We formalize the new domain in the abstract interpretation framework with abstract models of strings and objects commonly used in the existing JavaScript analyzers. For practical use of the domain, we present polynomial-time inclusion decision rules between the regular expressions and prove that the rules exactly capture the actual inclusion relation. We have implemented the domain as an extension of the open-source JavaScript analyzer, SAFE, and we show that the extension significantly improves the scalability and precision of the baseline analyzer in analyzing programs that use jQuery.

Compiling MATLAB---a dynamic, array-based language---to JavaScript is an attractive proposal: the output code can be deployed on a platform used by billions and can leverage the countless hours that have gone into making JavaScript JIT engines fast. But before that can happen, the original MATLAB code must be properly translated, making sure to bridge the semantic gaps of the two languages. An important area where MATLAB and JavaScript differ is in their handling of arrays: for example, in MATLAB, arrays are one-indexed and writing at an index beyond the end of an array extends it; in JavaScript, typed arrays are zero-indexed and writing out of bounds is a no-op. A MATLAB-to-JavaScript compiler must address these mismatches. Another salient and pervasive difference between the two languages is the assignment of arrays to variables: in MATLAB, this operation has value semantics, while in JavaScript is has reference semantics. In this paper, we present MatJuice --- a source-to-source, ahead-of-time compiler back-end for MATLAB --- and how it deals efficiently with this last issue. We present an intra-procedural data-flow analysis to track where each array variable may point to and which variables are possibly aliased. We also present the associated copy insertion transformation that uses the points-to information to insert explicit copies when necessary. The resulting JavaScript program respects the MATLAB value semantics and we show that it performs fewer run-time copies than some alternative approaches.

The RPython framework takes an interpreter for a dynamic language as its input and produces a Virtual Machine (VM) for that language. RPython is being used to develop PyPy, a high-performance Python interpreter. However, the produced VM does not support parallel execution since the framework relies on a Global Interpreter Lock (GIL): PyPy serialises the execution of multi-threaded Python programs. We describe the rationale and design of a new parallel execution model for RPython that allows the generation of parallel virtual machines while leaving the language semantics unchanged. This model then allows different implementations of concurrency control, and we discuss an implementation based on a GIL and an implementation based on Software Transactional Memory (STM). To evaluate the benefits of either choice, we adapt PyPy to work with both implementations (GIL and STM). The evaluation shows that PyPy with STM improves the runtime of a set of multi-threaded Python programs over PyPy with a GIL by factors in the range of 1.87x up to 5.96x when executing on a processor with 8 cores.

Mainstream programming languages like Java have limited support for language extensibility. Without mechanisms for syntactic abstraction, new programming styles can only be embedded in the form of libraries, limiting expressiveness. In this paper, we present Recaf, a lightweight tool for creating Java dialects; effectively extending Java with new language constructs and user defined semantics. The Recaf compiler generically transforms designated method bodies to code that is parameterized by a semantic factory (Object Algebra), defined in plain Java. The implementation of such a factory defines the desired runtime semantics. We applied our design to produce several examples from a diverse set of programming styles and two case studies: we define i) extensions for generators, asynchronous computations and asynchronous streams and ii) a Domain-Specific Language (DSL) for Parsing Expression Grammars (PEGs), in a few lines of code.

This paper introduces typy, a statically typed programming language embedded by reflection into Python. typy features a fragmentary semantics, i.e. it delegates semantic control over each term, drawn from Python's fixed concrete and abstract syntax, to some contextually relevant user-defined semantic fragment. The delegated fragment programmatically 1) typechecks the term (following a bidirectional protocol); and 2) assigns dynamic meaning to the term by computing a translation to Python. We argue that this design is expressive with examples of fragments that express the static and dynamic semantics of 1) functional records; 2) labeled sums (with nested pattern matching a la ML); 3) a variation on JavaScript's prototypal object system; and 4) typed foreign interfaces to Python and OpenCL. These semantic structures are, or would need to be, defined primitively in conventionally structured languages. We further argue that this design is compositionally well-behaved. It avoids the expression problem and the problems of grammar composition because the syntax is fixed. Moreover, programs are semantically stable under fragment composition (i.e. defining a new fragment will not change the meaning of existing program components.)

Nowadays many virtual execution environments benefit from concurrency offered by the actor model. Unfortunately while actors are used in many applications and computing frameworks existing profiling tools are little effective in analyzing the performance of applications using actors. In this paper we present a new instrumentation based technique to profile actors in virtual execution environments. Our technique adopts platform independent profiling metrics that minimize the perturbations induced by the instrumentation logic and allow comparing profiling results across different platforms. In particular our technique measures the initialization cost the amount of executed computations and the messages sent and received by each actor. We implement our technique within a profiling tool for Akka actors on the Java platform. Evaluation results show that our profiling technique can help performance analysis of actor utilization and communication between actors in large scale computing frameworks.

With increasing amounts of data available on the web and a diverse range of users interested in programmatically accessing that data, web automation must become easier. Automation helps users complete many tedious interactions, such as scraping data, completing forms, or transferring data between websites. However, writing web automation scripts typically requires an expert programmer because the writer must be able to reverse engineer the target webpage. We have built a record and replay tool, Ringer, that makes web automation accessible to non-coders. Ringer takes a user demonstration as input and creates a script that interacts with the page as a user would. This approach makes Ringer scripts more robust to webpage changes because user-facing interfaces remain relatively stable compared to the underlying webpage implementations. We evaluated our approach on benchmarks recorded on real webpages and found that it replayed 4x more benchmarks than a state-of-the-art replay tool.