pSather is a parallel extension of the existing object-oriented language Sather. It offers a shared-memory programming model which integrates both control- and dataparallel extensions. This integration increases the flexibility of the language to express different algorithms and data structures, especially on distributed-memory machines (e.g. CM-5). This report describes our design objectives and the programming language pSather in detail. ICSI and Eidgenossische Technische Hochschule (ETH), Zurich, Switzerland. E-mail: murer@icsi.berkeley.edu. y ICSI and Computer Science Division, U.C. Berkeley. E-mail: jfeldman@icsi.berkeley.edu. z ICSI and Computer Science Division, U.C. Berkeley. E-mail: clim@icsi.berkeley.edu. x ICSI E-mail: mseidel@icsi.berkeley.edu. ii Contents 1 Introduction 4 1.1 Roadmap of this Report : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 5 1.2 Grammar Notation : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 6 2...

When designing software, programmers usually think in terms of modules that are represented as functions and classes. But using existing configuration management systems, programmers have to deal with versions and configurations that are organized by files and directories. This is inconvenient and error-prone, since there is a gap between handling source code and managing configurations. We present a framework for programming environments that handles versions and configurations directly in terms of the functions and classes in source code. We show that with this framework, configuration management issues in software reuse and cooperative programming become easier. We also present a prototype environment that has been developed to verify our ideas. Keywords: Configuration management, Programming environment, Cooperative programming, Software reuse 1. Introduction Modern programming languages support constructs like functions and classes that let programmers decompose ...

Modern programming languages support constructs like functions and classes that let programmers decompose source programs into modules. However, existing programming environments do not allow programmers to handle configuration management directly with them. Instead, system building and version control are usually handled with different decomposition structures. The modules used in configuration management do not always match the modules in the source code. This is both inconvenient and error-prone, since there is a gap between handling the source code and managing the configurations. In this research we propose a framework for programming environments that handles configuration management directly in terms of the modules in the source code. We define the operations required for this purpose, study their semantics, and find a general strategy to support them. We show that with the ability to handle a large module as a single unit, software reuse and cooperative programming becomes easier. We also design and implement a prototype environment to verify our ideas. 1.

Abstract not found

1 2. INTRODUCTION 1 2.1. Portability and Purity 2 2.2. Goals of Borneo 3 2.3. Brief Description of an IEEE 754 Machine 3 2.4. Language Features for Floating Point Computation 6 3. FUTURE WORK 9 3.1. Incorporating Java 1.1 Features 9 3.2. Unicode Support 10 3.3. Flush to Zero 10 3.4. Variable Trapping Status 10 3.5. Parametric Polymorphism 10 4. CONCLUSION 10 5. ACKNOWLEDGMENTS 11 6. BORNEO LANGUAGE SPECIFICATION 13 6.1. indigenous 13 6.2. Floating Point Literals 16 6.3. Float, Double, and Indigenous classes 17 6.4. New Numeric Types 18 6.5. Floating Point System Properties 20 + This material is based upon work supported under a National Science Foundation Graduate Fellowship. Any opinions, findings, conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. ii 6.6. Fused mac 21 6.7. Rounding Modes 21 6.8. Floating Point Exception Handling 31 6.9. Operator Overloading 51 6.10...

. Concurrent languages have oered parallel loop constructs for some time to allow a parallel computation to be expressed in a simple and straightforward fashion. Modern programming languages include exceptions to allow for clean handling of errors or unexpected conditions, but few concurrent languages incorporate exception handling into their models for parallel loops. As a result, programmers that use parallel loops cannot use exceptions to simplify their programs. We present a semantics for handling exceptions in parallel loops that is predictable and that reduces to the familiar semantics for sequential loops. This semantics provides guarantees about the behavior of parallel loops even in the presence of exceptions, and facilitates the implementation of parallel algorithms. A Java library implementation of this semantics is presented, along with a description of a source-to-source translation. 1

Object-oriented programming has become quite widespread in recent years, although there are few guidelines to help us distinguish when a system is "truly" object-oriented or not. In this paper we discuss what have emerged as the main concepts in the object-oriented approach, and we attempt to motivate these concepts in terms of how they aid in software development.

email or url Abstract The abstract should be about 200 words, in a 10-point Roman italic font such as Times (preferred) or Computer Modern. It should summarise the scope and contribution of the paper in a concise and accurate manner without references to other works. The title should be centred in a 14-point Roman bold font followed by the authors names, affiliations, and either e-mail addresses or home page URL. Phone and fax numbers should be omitted.

The abstract should be about 200 words, in a 10point Roman italic font such as Computer Modern or Times. It should summarise the scope and results of the paper in a concise and accurate manner without references to other works. The title should be centred in a 14-point Roman bold font followed by the authors names, affiliations, and email addresses. Phone and fax numbers should be omitted. Keywords Computer science, database systems, computing theory, whatever. 1 Overall format Papers should be prepared for A4 paper, with 25 mm margins (equivalently, the text height should be 247 mm and the width 160 mm), using left and right justification, in a 10-point Roman font such as Computer Modern or Times, in two-column format. 1 The gutter between the columns should be 5 mm wide. No page headers, footers or numbers should be included. Citations should be in the IEEE style, for example Hoare [4], or Ganapathi and Fischer [3], or Barnes [1, 2]. Authors should number the pages by writing on...

The abstract should be about 200 words, in a 10point Roman italic font such as Computer Modern or Times. It should summarise the scope and contribution of the paper in a concise and accurate manner without references to other works. The title should be centred in a 14-point Roman bold font followed by the authors names, affiliations, and email addresses. Phone and fax numbers should be omitted. Keywords A list of keywords based on those given in the Call for Papers should be included. 1 Overall format Papers should be prepared for A4 paper, with 25 mm margins (equivalently, the text height should be 247 mm and the width 160 mm), using left and right justification, in a 10-point Roman font such as Computer Modern or Times, in two-column format. The gutter between the columns should be 5 mm wide. No page headers, footers or numbers should be included. Citations should be in the IEEE style, for example [3], or [1, 2]. Do not include page numbers in your paper. Authors submitting printe...