Optimizing interpreters by tuning opcode orderings on virtual machines for modern architectures

McCandless, Jason; Gregg, David

doi:10.1145/2093157.2093183

Cited by 5 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lot of techniques, like feedback-guided technique [22], profile-guided technique [23], etc, are conducted to achieve better orderings, improving code locality and reducing cache misses. Jason McCandless and his co-worker [24] implement a metaheuristic (Monte Carlo) to generate better orderings, achieving considerable performance improvement. As shown in section 5, cache misses can also be reduced by our new mechanism, making the new interpreter perform better.…”

Section: Related Workmentioning

confidence: 99%

Performance Research and Optimization on CPython’s Interpreter

Cao¹,

Gu²,

Ren³

et al. 2015

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-In this paper, the performance research on CPython's latest interpreter is presented, concluding that bytecode dispatching takes about 25 percent of total execution time on average. Based on this observation, a novel bytecode dispatching mechanism is proposed to reduce the time spent on this phase to a minimum. With this mechanism, the blocks associated with each kind of bytecodes are rewritten in hand-tuned assembly, their opcodes are renumbered, and their memory spaces are rescheduled. With these preparations, this new bytecode dispatching mechanism replaces the time-consuming memory reading operations with rapid operations on registers.This mechanism is implemented in CPython-3.3.0. Experiments on lots of benchmarks demonstrate its correctness and efficiency. The comparison between original CPython and optimized CPython shows that this new mechanism achieves about 8.5 percent performance improvement on average. For some particular benchmarks, the maximum improvement is up to 18 percentages.

show abstract

Section: Related Workmentioning

confidence: 99%

Performance Research and Optimization on CPython’s Interpreter

Cao¹,

Gu²,

Ren³

et al. 2015

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

show abstract

“…Finally, modifying the interpreter directly to turn features on and off would be both difficult and possibly plagued by second-order effects due to orthogonal issues such as the ordering of bytecode implementations in the interpreter [MG11].…”

Section: Compiling Python Functions With Pylibjitmentioning

confidence: 99%

Python Interpreter Performance Deconstructed

Barany

2014

Proceedings of the Workshop on Dynamic Languages and Applications

View full text Add to dashboard Cite

The Python programming language is known for performing poorly on many tasks. While to some extent this is to be expected from a dynamic language, it is not clear how much each dynamic feature contributes to the costs of interpreting Python. In this study we attempt to quantify the costs of language features such as dynamic typing, reference counting for memory management, boxing of numbers, and late binding of function calls.We use an experimental compilation framework for Python that can make use of type annotations provided by the user to specialize the program as well as elide unnecessary reference counting operations and function lookups. The compiled programs run within the Python interpreter and use its internal API to implement language semantics. By separately enabling and disabling compiler optimizations, we can thus measure how much each language feature contributes to total execution time in the interpreter.We find that a boxed representation of numbers as heap objects is the single most costly language feature on numeric codes, accounting for up to 43 % of total execution time in our benchmark set. On symbolic object-oriented code, late binding of function and method calls costs up to 30 %. Redundant reference counting, dynamic type checks, and Python's elaborate function calling convention have comparatively smaller costs.Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org.Python is a popular programming language, but it has a reputation of being slow. Projects such as the PyPy just-intime compiler [BCFR09] show that it is possible to execute some Python programs up to 50 times faster while remaining faithful to its semantics.Why, then, is the standard Python interpreter (often called CPython) so slow? Is it due to interpretation overhead? Due to dynamic typing and late binding of operations? Due to the boxing of numbers, or due to the overhead of automatic memory management by reference counting?We attempt to answer some of these questions by measuring the execution times of Python programs with and without enabling the corresponding language features. Simply turning off language features is obviously infeasible in the standard interpreter, but we sidestep this problem in a novel way by using a compiler to specialize programs while still running them inside the interpreter. Type annotations from the user and simple program analysis allow us to turn off language features where it is safe to do so, but to still adhere to Python's dynamic semantics otherwise.This kind of analysi...

show abstract

Improved Branch Prediction for Just-in-Time Decompression of Canonical Huffman Bytecode Streams

Jeong

Burgstaller

2014

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Optimizing interpreters by tuning opcode orderings on virtual machines for modern architectures

Cited by 5 publications

References 20 publications

Performance Research and Optimization on CPython’s Interpreter

Performance Research and Optimization on CPython’s Interpreter

Python Interpreter Performance Deconstructed

Improved Branch Prediction for Just-in-Time Decompression of Canonical Huffman Bytecode Streams

Contact Info

Product

Resources

About