The concept of “threaded code” is presented as an alternative to machine language code. Hardware and software realizations of it are given. In software it is realized as interpretive code not needing an interpreter. Extensions and optimizations are mentioned.

There is an increasing trend that intermediate representations (IRs) are used to deliver programs in more and more languages, such as Java. Although Java can provide many advantages, including a wider portability and better optimisation opportunities on execution, it introduces extra overhead by requiring an IR translation for the program execution. For maximum execution performance, an optimising compiler is placed in the runtime to selectively optimise code regions regarded as “hotspots”. This common approach has been effectively deployed in many implementation of programming languages. However, the computational resources demanded by this approach made it less efficient, or even difficult to deploy directly in a resourceconstrained environment. One implementation approach is to use a remote compilation technique to support compilation during the execution. The work presented in this dissertation supports the thesis that execution performance can be improved by the use of efficient optimising compilation by using a proxy dynamic optimising compiler. After surveying various approaches to the design and implementation of remote compilation, a proxy compilation system called Apus is defined. To demonstrate the effectiveness of using a dynamic optimising compiler as a proxy compiler, a complete proxy compilation system is written based on a research-oriented Java VirtualMachine (JVM). The proxy compilation system is discussed in detail, showing how to deliver remote binaries and manage a cache of binaries by using a code migration approach. The proxy compilation client shows how the proxy compilation service is integrated with the selective optimisation system to maximise execution performance. The results of empirical measurements of the system are given, showing the efficiency of code optimisation from either the proxy compilation service and a local binary cache. The conclusion of this work is that Java execution performance can be improved by efficient optimising compilation with a proxy compilation service by using a code migration technique.

Threaded code

Proxy compilation for Java via a code migration technique

As the backbone of many computing environments today, it is important that process virtual machines be both performant and robust in mobile, personal desktop, and enterprise applications. This thesis focusses on code generation within these virtual machines, particularly addressing situations where redundant work is being performed. The goal is to exploit information sharing in order to improve the performance and robustness of virtual machines that are accelerated by native code generation. First, the thesis investigates the potential to share generated code between multiple threads in a dynamic binary translator used to perform instruction set simulation. This is done through a code generation design that allows native code to be executed by any simulated core and adding a mechanism to share native code regions between threads. This is shown to improve the average performance of multi-threaded benchmarks by 1.4x when simulating 128 cores on a quad-core host machine. Secondly, the ahead-of-time code generation system used for executing Android applications is improved through the use of profiling. The thesis investigates the potential for profiles produced by individual users of applications to be shared and merged together to produce a generic profile that still provides a lot of benefit for a new user who is then able to skip the expensive profiling phase. These profiles can not only be used for selective compilation to reduce code-size and installation time, but can also be used for focussed optimisation on vital code regions of an application in order to improve overall performance. With selective compilation applied to a set of popular Android applications, code-size can be reduced by 49.9% on average, while installation time can be reduced by 31.8%, with only an average 8.5% increase in the amount of sequential runtime required to execute the collected profiles. The thesis also shows that, among the tested users, the use of a crowd-sourced and merged profile does not significantly affect their estimated performance loss from selective compilation (0.90x-0.92x) in comparison to when they they perform selective compilation with their own unique profile (0.93x). Furthermore, by proposing a new, more powerful code generator for Android’s virtual machine, these same profiles can be used to perform focussed optimisation, which preliminary results show to increase runtime performance across a set of common Android benchmarks by 1.46x-10.83x. Finally, in such a situation where a new code generator is being added to a virtual machine, it is also important to test the code generator for correctness and robustness. The methods of execution of a virtual machine, such as interpreters and code generators, must share a set of semantics about how programs must be executed, and this can be exploited in order to improve testing. This is done through the application of domain-aware binary fuzzing and differential testing within Android’s virtual machine. The thesis highlights a series of actual code generation and verification bugs that were found in Android’s virtual machine using this testing methodology, as well as comparing the proposed approach to other state-of-the-art fuzzing techniques.

Applications of information sharing for code generation in process virtual machines

Automatic generation of optimised virtual machine interpreters

Runtime compilation of runtime-constructed code is becoming standard practice in libraries, DSLs, and database management systems. Since compilation is expensive, systems that are sensitive to compile times such as relational database query compilers compile only hot code and interprets the rest with a much slower interpreter. 
We present a code generation technique that lowers an AST to binary code by stitching together code from a large library of binary AST node implementations. We call the implementations stencils because they have holes where values must be inserted during code generation. We show how to construct such a stencil library and describe the copy-and-patch technique that generates optimized binary code. 
The result is a code generator with negligible cost: it produces code from an AST in less time than it takes to construct the AST. Compared to LLVM, compilation is two orders of magnitude faster than -O0 and three orders of magnitude faster than higher optimization levels. The generated code runs an order of magnitude faster than interpretation and runs even faster than LLVM -O0. Thus, copy-and-patch can effectively replace both interpreters and LLVM -O0, making code generation more effective in compile-time sensitive applications.

Copy-and-Patch Binary Code Generation

Interpreters play an important role in many languages, and their performance is critical particularly for the popular language Java. The performance of the interpreter is important even for high-pe...

Threaded code

Citations

Proxy compilation for Java via a code migration technique

Applications of information sharing for code generation in process virtual machines

Automatic generation of optimised virtual machine interpreters

Copy-and-Patch Binary Code Generation

Bytecode fetch optimization for a Java interpreter

Related Papers (5)

Optimizing direct threaded code by selective inlining

Optimizing an ANSI C interpreter with superoperators

Shade: a fast instruction-set simulator for execution profiling

Efficient implementation of the smalltalk-80 system

A brief history of just-in-time