TurboJ, a Java bytecode-to-native compiler

Weiß, Michael; Ferrière, François de; Delsart, Bertrand; Fabre, Christian; Hirsch, Frederick; Johnson, Ethan; Joloboff, Vania; Roy, Fred; Siebert, Fridtjof; Spengler, Xavier

doi:10.1007/bfb0057785

Cited by 19 publications

(22 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Standalonemode AOTCs, like Toba [4] and GCJ [2], translate the whole application to native code as a standalone executable. On the other hand, mixed-mode AOTCs only compile the hot spots and interact with VM, such as Harissa [5], TurboJ [6] and our work. Despite being able to use the AOTC to translate bytecode into native code directly, we take the path to generate C code first and then compile the C code with existing compilers to exploit the advantages of mature machine independent and machine dependent optimizations.…”

Section: Related Workmentioning

confidence: 99%

A method-based ahead-of-time compiler for android applications

Wang

Pérez

Chung

et al. 2011

Proceedings of the 14th International Conference on Compilers, Architectures and Synthesis for Embedded Systems

View full text Add to dashboard Cite

The execution environment of Android system is based on a virtual machine called Dalvik virtual machine (DVM) in which the execution of an application program is in interpret-mode. To reduce the interpretation overhead of DVM, Google has included a trace-based just-in-time compiler (JITC) in the latest version of Android. Due to limited resources and the requirement for reasonable response time, the JITC is unable to apply deep optimizations to generate high quality code. In this paper, we propose a method-based ahead-of-time compiler (AOTC), called Icing, to speed up the execution of Android applications without the modification of any components of Android framework. The main idea of Icing is to convert the hot methods of an application program from DEX code to C code and uses the GCC compiler to translate the C code to the corresponding native code. With the Java Native Interface (JNI) library, the translated native code can be called by DVM. Both AOTC and JITC have their strength and weakness. In order to combine the strength and avoid the weakness of AOTC and JITC, in Icing, we have proposed a cost model to determine whether a method should be handled by AOTC or JITC during profiling. To evaluate the performance of Icing, four benchmarks used by Google JITC are used as test cases. The performance results show that, with Icing, the execution time of an application is two to three times faster than that without JITC, and 25% to 110% faster than that with JITC.

show abstract

Section: Related Workmentioning

confidence: 99%

A method-based ahead-of-time compiler for android applications

Wang

Pérez

Chung

et al. 2011

Proceedings of the 14th International Conference on Compilers, Architectures and Synthesis for Embedded Systems

View full text Add to dashboard Cite

show abstract

“…We have developed two versions of this library: one POSIX compliant (which works unchanged in both FAST-OS and eCos) that we have described herein, and a non-POSIX one that uses OS alarms and alarm handlers directly (for eCos). Our approach has been first integrated in an industry-strength RTSJ-compliant compilation infrastructure and run-time environment [28]: the model extraction and synthesis steps were interfaced with the Java-to-C TurboJ compilation chain [44] and our controller subsystem was part of the Expresso executive [16]. More recently, it was used as part of an MDE framework for real-time embedded systems comprising the formal, modeltransformation and code-generation tool Jahuel [5] and STMicroelectronics's FlexCC2 compilation technology [7].…”

Section: Discussionmentioning

confidence: 99%

A model-based approach for multiple QoS in scheduling: from models to implementation

Kloukinas

Yovine

2010

Autom Softw Eng

View full text Add to dashboard Cite

This is the unspecified version of the paper.This version of the publication may differ from the final published version. Abstract Meeting multiple Quality of Service (QoS) requirements is now an important factor for the success of complex software systems. This paper presents a correct-byconstruction, automated, model-driven scheduler synthesis approach for scheduling system tasks so as to meet multiple QoS requirements. As a first step, it shows how software engineers can meet deadlock-freedom and timeliness requirements, in a manner that (i) does not over-provision resources, (ii) does not require architectural changes to the system, and that (iii) leaves enough degrees of freedom to pursue further properties. The synthesis methodology directly associates each scheduler with a specific pair of QoS property and underlying platform execution model, so as to facilitate their validation and the understanding of the overall system behaviour, required to meet further QoS properties. The paper shows how the methodology is applied in practice and also presents the implementation infrastructure needed for executing an application on top of common operating systems, without requiring modifications of the operating system. Permanent repository link

show abstract

“…For example, standard Sun Java Virtual Machine (JVM) needs at least several megabytes of memory just to start its execution. Usually one has to employ special approaches such as customizable JVMs with small memory-footprint [4] [5], or ahead-of-time compilation [6]. This special approaches, however, demand additional effort.…”

Section: A Programming Languages Abstractionmentioning

confidence: 99%