A nanopass infrastructure for compiler education

Sarkar, Dipanwita; Waddell, Oscar; Dybvig, R. Kent

doi:10.1145/1016850.1016878

Cited by 17 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To compile this high level language, we first translate it to an intermediate language similar to C, in which the order of evaluation is explicit in its syntax. The compiler is written in a nanopass style [Sarkar et al 2004], and is made up of five passes: (1) typecheck type checks the source program, (2) uniqify freshens all bound variables to handle shadowing, (3) explicateControl translates to an intermediate language similar to C, (4) selectInstructions generates x86 code which has variables in it, ( 5) and assignHomes maps each variable to a location on the stack. The input to this benchmark is a synthetically generated, balanced syntax-tree with conditional expressions at the top, followed by a sequence of let bindings.…”

Section: Results: X86-compiler Case Studymentioning

confidence: 99%

Efficient Tree-Traversals: Reconciling Parallelism and Dense Data Representations

Koparkar¹,

Rainey²,

Vollmer³

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent work showed that compiling functional programs to use dense, serialized memory representations for recursive algebraic datatypes can yield significant constant-factor speedups for sequential programs. But serializing data in a maximally dense format consequently serializes the processing of that data, yielding a tension between density and parallelism. This paper shows that a disciplined, practical compromise is possible. We present Parallel Gibbon, a compiler that obtains the benefits of dense data formats and parallelism. We formalize the semantics of the parallel location calculus underpinning this novel implementation strategy, and show that it is type-safe. Parallel Gibbon exceeds the parallel performance of existing compilers for purely functional programs that use recursive algebraic datatypes, including, notably, abstract-syntax-tree traversals as in compilers. * Extended version of "Efficient Tree-Traversals: Reconciling Parallelism and Dense Data Representations, " [Koparkar et al.

show abstract

Section: Results: X86-compiler Case Studymentioning

confidence: 99%

Efficient Tree-Traversals: Reconciling Parallelism and Dense Data Representations

Koparkar¹,

Rainey²,

Vollmer³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Its authors developed it while implementing a nanopass-based compiler [Sarkar et al 2004] and developed it in order for their "compiler passes to traverse the abstract syntax tree quickly." Though we have chosen a different approach, our system is motivated by similar concerns.…”

Section: Alloymentioning

confidence: 99%

Template your boilerplate

Adams

Dubuisson

2012

SIGPLAN Not.

View full text Add to dashboard Cite

Generic programming allows the concise expression of algorithms that would otherwise require large amounts of handwritten code. A number of such systems have been developed over the years, but a common drawback of these systems is poor runtime performance relative to handwritten, non-generic code. Generic-programming systems vary significantly in this regard, but few consistently match the performance of handwritten code. This poses a dilemma for developers. Generic-programming systems offer concision but cost performance. Handwritten code offers performance but costs concision.This paper explores the use of Template Haskell to achieve the best of both worlds. It presents a generic-programming system for Haskell that provides both the concision of other genericprogramming systems and the efficiency of handwritten code. Our system gives the programmer a high-level, generic-programming interface, but uses Template Haskell to generate efficient, nongeneric code that outperforms existing generic-programming systems for Haskell. This paper presents the results of benchmarking our system against both handwritten code and several other genericprogramming systems. In these benchmarks, our system matches the performance of handwritten code while other systems average anywhere from two to twenty times slower.

show abstract

“…Sarkar et al [23] developed a compiler course (using Scheme) based on the nanopass framework, where the compiler consists of many small translation (or verification) processes written in a domain specific language developed for this purpose. Unlike nanopass, we chose to use ordinary ML as the meta language in order to avoid the overhead of understanding such a domain specific language itself, and to utilize the type system of ML for statically checking the syntactic validity of intermediate code even before running the compiler.…”

Section: Related Workmentioning

confidence: 99%

MinCaml

Sumii

2005

Proceedings of the 2005 Workshop on Functional and Declarative Programming in Education

View full text Add to dashboard Cite

We present a simple compiler, consisting of only 2000 lines of ML, for a strict, impure, monomorphic, and higher-order functional language. Although this language is minimal, our compiler generates as fast code as standard compilers like Objective Caml and GCC for several applications including ray tracing, written in the optimal style of each language implementation. Our primary purpose is education at undergraduate level to convince students-as well as average programmers-that functional languages are simple and efficient.

show abstract

A nanopass infrastructure for compiler education

Cited by 17 publications

References 12 publications

Efficient Tree-Traversals: Reconciling Parallelism and Dense Data Representations

Efficient Tree-Traversals: Reconciling Parallelism and Dense Data Representations

Template your boilerplate

MinCaml

Contact Info

Product

Resources

About