DOCSLIB.ORG

Lecture 7: Binding Time and Storage

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Lecture 7: Binding Time and Storage Lecture 7: Binding Time and Storage COMP 524 Programming Language Concepts Stephen Olivier February 5, 2009 Based on notes by A. Block, N. Fisher, F. Hernandez-Campos, and D. Stotts The University of North Carolina at Chapel Hill Goal of Lecture •The Goal of this lecture is to discuss object binding and memory management. The University of North Carolina at Chapel Hill 2 High-Level Programming Languages •High-Level Programming languages are defined by two characteristics • Machine “independence” • Ease of programming The University of North Carolina at Chapel Hill 3 Machine “Independence” •With few exceptions, the code of a high-level language can be compiled on any system • For example cout << “hello world”<< endl; means the same thing on any machine •However, few languages are completely machine independent. •Generally, the more machine dependent a language is the more “efficient” it is. The University of North Carolina at Chapel Hill 4 Ease of Programming Names Control Flow Types Subroutines Object Orientation Concurrency Declarative Programming The University of North Carolina at Chapel Hill 5 Naming •Naming is the process by which a programer associates a name with a potentially complicated program fragment. • Purpose is to hide complexity. • For example, to designate variables, types, classes, operators, etc ... The University of North Carolina at Chapel Hill 6 Naming •Naming is the process by which a programer associates a name with a potentially complicated program fragment. By naming an • Purpose is to hide complexity. object we • For example, to designate variables, types, classes, operators, etc ... make an abstraction The University of North Carolina at Chapel Hill 7 Abstractions •Control abstractions allows programs to “hide” complex code behind simple interface • Subroutines and functions • Classes. •Data abstraction allow the programer to hide data representation details behind abstract operations • Abstract Data Types (ADTs) • Classes. The University of North Carolina at Chapel Hill 8 Binding Time •A binding is an association between any two things • Name of an object and the object. • A Dook student to a loosing basketball team. •Binding Time is the time at which a binding is created. The University of North Carolina at Chapel Hill 9 Binding Time Language Design Time Language Implementation Time Increasing Efficiency Increasing Program Writing Time Compile Time Link Time Increasing Flexibility Increasing Load Time Run Time The University of North Carolina at Chapel Hill 10 Object Lifetime •Object lifetimes have two components • Lifetime of the object. • Lifetime of the binding. •These two don’t necessarily correspond. • For example in C++, when a variable is passed by “reference”, i.e., using “&”, then the name of the object does not exist even though the binding does. • For example in C++, when the value pointed to by an object is deleted the binding is gone before the object. The University of North Carolina at Chapel Hill 11 Object Lifetimes •Object Lifetimes correspond to three principal storage allocation mechanisms, • Static objects, which have an absolute address • Stack objects, which are allocated and deallocated in a Last- In First-Out (LIFO) order • Heap objects, which are allocated and deallocated at arbitrary times. The University of North Carolina at Chapel Hill 12 Static Allocation •Under static allocation, objects are given an absolute address that is retained through the program’s execution • e.g., global variables Temps. Temps. Temps. Local Variables Local Variables Local Variables Misc. Bookkeeping Misc. Misc. Bookkeeping Bookkeeping Return Address Return Address Return Address Arguments & Arguments & Arguments & Returns Returns Returns SubroutineThe University 1 of NorthSubroutine Carolina at2 Chapel Hill Subroutine X 13 Stack Allocation •Under stack-based allocation, objects are allocated in a Last-In First-Out (LIFO) basis called a stack. • e.g., recursive subroutine parameters. sp Args to called routines. Subroutine D fp Temps Subroutine C Local Variables Misc. Subroutine B Bookkeeping fp Subroutine A Return Address (called from main) TheStack growth University of North Carolina at Chapel Hill 14 fp is the “Frame Pointer” Stack Allocation •Under stack-basedsp is the allocation “Stack ,Pointer objects are” allocated in a Last-In First-Out (LIFO) basis called a stack. • e.g., recursive subroutine parameters. sp Args to called routines. Subroutine D fp Temps Subroutine C Local Variables Misc. Subroutine B Bookkeeping fp Subroutine A Return Address (called from main) TheStack growth University of North Carolina at Chapel Hill 15 Calling Sequence •On procedure call and return compilers generate code that execute to manage the runtime stack. • Setup at call to procedure foo(a,b). • Prologue before foo code executes. • Epilogue at the end of foo code. • “Teardown” right after calling the code. The University of North Carolina at Chapel Hill 16 Setup foo(a,b) •Move sp to allocate a new stack frame •Copy args a,b into frame •Copy return address into frame •Set fp to point to new frame •Maintain static chain or display •Move PC to procedure address The University of North Carolina at Chapel Hill 17 Setup foo(a,b) •Move sp to allocate a new stack frame •Copy args x,y into frame •Copy return address into sp frame •Set fp to point to new frame Subroutine Bar fp •Maintain static chain or Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 18 Setup foo(a,b) •Move sp to allocate a new stack frame •Copy args x,y into frame sp Subroutine foo •Copy return address into frame •Set fp to point to new frame Subroutine Bar fp •Maintain static chain or Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 19 Setup foo(a,b) •Move sp to allocate a new stack frame •Copy args x,y into frame sp Subroutine foo •Copy return address into frame x & y •Set fp to point to new frame Subroutine Bar fp •Maintain static chain or Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 20 Setup foo(a,b) •Move sp to allocate a new stack frame sp •Copy args x,y into frame Subroutine foo fp •Copy return address into Return Ad frame x & y •Set fp to point to new frame Subroutine Bar •Maintain static chain or Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 21 Setup foo(a,b) •Move sp to allocate a new stack frame sp •Copy args x,y into frame Subroutine foo fp •Copy return address into Return Ad frame x & y •Set fp to point to new frame Subroutine Bar •Maintain static chain or Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 22 Setup foo(a,b) •Move sp to allocate a new stack frame sp •Copy args x,y into frame Subroutine foo fp •Copy return address into Return Ad frame A & B •Set fp to point to new frame We’llSubroutine ignore Bar this for now •Maintain static chain or Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 23 Setup foo(a,b) •Move sp to allocate a new stack frame sp •Copy args x,y into frame Subroutine foo fp •Copy return address into Return Ad frame This changesA & B •Set fp to point to new frame whereSubroutine the code Bar •Maintain static chain or is executed.Subroutine B display Subroutine A (called from main) •Move PC to procedure address The University of North Carolina at Chapel Hill 24 Prologue •Copy registers into local slots •Object initialization. The University of North Carolina at Chapel Hill 25 Prologue •Copy registers into local slots •Object initialization. Return Ad fp x & y Subroutine Bar The University of North Carolina at Chapel Hill 26 Prologue •Copy registers into local slots •Object initialization. All old reg. Old fp Return Ad fp x & y Subroutine Bar The University of North Carolina at Chapel Hill 27 Prologue •Copy registers into local slots •Object initialization. Objects that areAll old used reg. are initialized.Old fp Return Ad fp x & y Subroutine Bar The University of North Carolina at Chapel Hill 28 Epilogue •Place return value into slot in frame. sp •Restore registers. Subroutine foo fp •Restore PC to return Return Ad address. x & y Subroutine Bar Subroutine B Subroutine A (called from main) The University of North Carolina at Chapel Hill 29 Epilogue •Place return value into slot in frame. sp •Restore registers. Subroutine foo fp •Restore PC to return Return Ad address. x & y Subroutine Bar Return Value Subroutine B Subroutine A (called from main) The University of North Carolina at Chapel Hill 30 Epilogue Registers stored from “foo”’s subroutine are •Place return value into slot registered. in frame. sp •Restore registers. Subroutine foo fp •Restore PC to return Return Ad address. x & y Subroutine Bar Return Value Subroutine B Subroutine A (called from main) The University of North Carolina at Chapel Hill 31 Epilogue The program resumes from where it began. •Place return value into slot in frame. sp •Restore registers. Subroutine foo fp •Restore PC to return Return Ad address. x & y Subroutine Bar Return Value Subroutine B Subroutine A (called from main) The University of North Carolina at Chapel Hill 32 “Teardown” •Move sp & fp (deallocate frame) sp •Move return values (if in Subroutine foo fp registers) Return Ad x & y Subroutine Bar Return Value Subroutine B Subroutine A (called from main) The University of North Carolina at Chapel Hill 33 “Teardown” •Move sp & fp (deallocate frame) •Move return values (if in registers) sp Subroutine Bar Return Value fp Subroutine B Subroutine A (called from main) The University of North Carolina at Chapel Hill 34 “Teardown” If the return value was placed in a register, put •Move sp & fp (deallocate it in the stack.
Load more

Recommended publications
  • CS 403/503: Programming Languages
    BNF for <expression> <expression> → identifier | number | CS 403 -<expression> | (<expression>) | <expression><operator><expression> Organization of Programming Languages <operator> → + | - | * | / Class 3 - August 30, 2001 Topics For Today Parse Tree Review of BNF (Chapter 2.1) Derivations and parse trees (Chapter 2.1) Binding time (Chapter 3.1) Object lifetime and storage slope * x + intercept management (Chapter 3.2) BNF Concept Review Another Parse Tree Terminal symbols: Actual items that are part of the language (0,1,+,-,*,/,etc.) Non-Terminal symbols: <expr>, <op>, etc. Rule/production: A single line of BNF Alternatives: Expressed with | Also: slope * x + intercept *: 0 or more occurrences +:1 or more occurrences 1 Binding Lifetimes • Binding lifetime: The period of time Def: A binding is an association, such between the creation and destruction of a as between an attribute and an entity, name-to-object binding • Object lifetime: The period of time or between an operation and a between the creation and destruction of symbol. an object • These don’t necessarily coincide Def: Binding time is the time at which a • EX. Reference parameters binding takes place. Possible binding times Object Lifetimes… 1. Language design time--e.g., bind operator …correspond to one of three principal symbols to operations 2. Language implementation time--e.g., bind fl. pt. storage allocation mechanisms: type to a representation Static objects: Object lifetime = program 3. Compile time--e.g., bind a variable to a type in C execution. Object bound to one storage or Java location from load time on. 4. Load time--e.g., bind a FORTRAN 77 variable to a memory cell (or a C static variable) Stack objects: Object lifetime = subroutine 5.
    [Show full text]
  • C++/CLI Language Specification
    Ecma/TC39-TG5/2004/25 C++/CLI Language Specification Working Draft 1.5, Jun, 2004 Public Review Document Text highlighted like this indicates a placeholder for some future action. It might be a note from the editor to himself, or an indication of the actual or expected direction of some as-yet open issue. Note: In the spirit of the "Working Draft, Standard for Programming Language C++", this is an early draft. It’s known to be incomplet and incorrekt, and it has lots of bad formatting. Publication Time: 6/17/2004 11:44 PM Table of Contents Table of Contents Introduction....................................................................................................................................................xi 1. Scope............................................................................................................................................................. 1 2. Conformance ............................................................................................................................................... 2 3. Normative references .................................................................................................................................. 3 4. Definitions .................................................................................................................................................... 4 5. Notational conventions................................................................................................................................ 7 6. Acronyms and abbreviations
    [Show full text]
  • C# Programming in Depth Lecture 4: Garbage Collection & Exception
    Chair of Software Engineering C# Programming in Depth Prof. Dr. Bertrand Meyer March 2007 – May 2007 Lecture 4: Garbage Collection & Exception Handling Lisa (Ling) Liu Overview Scope and lifetime Garbage collection mechanism Exception handling C# programming lecture 4: Garbage Collection & Exception Handling 2 Scope and lifetime Scope of a variable is portion of program text within which it is declared ¾ Need not be contiguous ¾ In C#, is static: independent of data Lifetime or extent of storage is portion of program execution during which it exists ¾ Always contiguous ¾ Generally dynamic: dependent on data Class of lifetime ¾ Static: entire duration of program ¾ Local or automatic: duration of call or block execution (local variable) ¾ Dynamic: From time of allocation statement (new) to deallocation, if any. C# programming lecture 4: Garbage Collection & Exception Handling 3 Object lifetime in C# Memory allocation for an object should be made using the “new” keyword Objects are allocated onto the managed heap, where they are automatically deallocated by the runtime at “some time in the future” Garbage collection is automated in C# Rule: Allocate an object onto the managed heap using the new keyword and forget about it C# programming lecture 4: Garbage Collection & Exception Handling 4 Object creation When a call to new is made, it creates a CIL “newobj” instruction to the code module public static int Main (string[] args) { Car c = new Car(“Viper”, 200, 100); } IL_000c: newobj instance void CilNew.Car::.ctor (string, int32,
    [Show full text]
  • Generating Object Lifetime Traces with Merlin
    Generating Object Lifetime Traces with Merlin MATTHEW HERTZ University of Massachusetts, Amherst STEPHEN M BLACKBURN Australian National University J ELIOT B MOSS University of Massachusetts, Amherst KATHRYN S McKINLEY University of Texas at Austin and DARKO STEFANOVIC´ University of New Mexico ÈÖÓÖÑÑÖ× Ö ÛÖØÒ ÖÔÐÝ ÖÓÛÒ ÒÙÑ Ö Ó ÔÖÓÖÑ× Ò Ó Ø ÐÒÙ׸ × ÂÚ Ò ¸ ØØ ÖÕÙÖ Ö Ö Ò ×ÑÙÐØÓÒ ×Ô ÜÔÐÓÖØÓÒ ÙÔ Ý ÒÐÒ Ô Ö ÙÒÖ×ØÒÒ× Ó Ó ÐØÑ ÚÓÖ Ò Ò ×Ò Ó ÒÛ Ö ÐÓÖØÑ׺ ÏÒ ÒÖØÒ Ô Ø ÖÙعÓÖ ÑØÓ Ó Ó ÐØÑ× ÖÕÙÖ× ÛÓÐ¹Ô Ö Ø ÚÖÝ Ô ÓØÒØÐ Ô ÓÒØ Ò Ø ÔÖÓÖѺ Ø× ÔÖÓ × ÔÖÓØÚÐÝ ÜÔ Ò×Ú¸ Ö× ÓØÒ ÒÙÐØ Ý ÓÒÐÝ Ô ÖÓ ¸ ºº¸ ÚÖÝ ¿¾Ã ÝØ× Ó ÐÐÓ Ù× Ö Û Ï ÜØÒ Ø ×ØØ Ó Ø ÖØ ÓÖ ×ÑÙÐØÒ Ö ÐÓÖØÑ× Ò ØÛÓÛÝ׺ Ö×ظ ÚÐÓÔ ÑØÓ ÓÐÓÝ ÓÖ ×ÑÙÐØÓÒ ×ØÙ× Ó ÝÒ Ö Ò ÔÖ×ÒØ Ö×ÙÐØ× ×ÓÛÒ Ø Ó ÖÒÙÐÖØÝ ÓÒ Ø× ×ÑÙÐØÓÒ׺ Ï ×ÓÛ ØØ ÖÒÙÐÖØÝ ÓØÒ ×ØÓÖØ× ×ÑÙÐØ Ö Ö×ÙÐØ× ÛØ Ô Û ÓÖ Ó ÔÖ×ÒØ Ò Ñ×ÙÖ Ø Ô Ó ÒÛ ÐÓÖØÑ ÅÖÐÒ ÐØÑ׺ ÅÖÐÒ ØÑ×ØÑÔ× Ó Ò ÐØÖ Ù×× Ø ØÑ×ØÑÔ× Ó Ó ØÓ ÛÒ ØÝ º Ì ÅÖÐÒ ÐÓÖØÑ × ÓÒ Ö Ô ÖÓÖÑ Ý Ø × ÖÑÒØÐ Ö×ÙÐØ× ×ÓÛ ØØ ÅÖÐÒ ÒÖØ ÓÚÖ ØÛÓ ÓÖÖ× Ó ÑÒØÙ ×Ý×ØѺ ÜÔ ×ØÖ ØÒ Ø ÑØÓ ØÖ ÚÖÝ Ó ÐÐÓ Ï Ð×Ó Ù× ÅÖÐÒ ØÓ ÔÖÓ Ú×ÙÐÞØÓÒ× Ó Ô ÚÓÖ ØØ ÜÔ Ó× ÒÛ Ó ÐØÑ ÚÓÖ׺ Categories and Subject Descriptors: D.3.4 [Programming Languages]: Processors—memory management (gar- bage collection) ÒÖÐ ÌÖÑ× ÐÓÖØÑ׸ ÄÒÙ׸ È ØÓÒÐ ÃÝ ÏÓÖ× Ò ÈÖ×× Ö ×Ò¸ Ó ÐØÑ ÒÐÝ×׸ ÒÖØÓÒ This work is supported by NSF grants CCR-0219587, ITR CCR-0085792, EIA-0218262, EIA-9726401, EIA- 030609, EIA-0238027, EIA-0324845, DARPA F33615-03-C-4106, Hewlett-Packard gift 88425.1, Microsoft Re- search, and IBM.
    [Show full text]
  • Modern C++ Object-Oriented Programming
    CPP Modern C++ Object-Oriented Programming ''Combine old and newer features to get the best out of the language'' Margit ANTAL 2020 C++ - Object-Oriented Programming Course content − Introduction to C++ − Object-oriented programming − Generic programming and the STL − Object-oriented design C++ - Object-Oriented Programming References − Bjarne Stroustrup, Herb Sutter, C++ Core Guidelines, 2017. rd − M. Gregoire, Professional C++, 3 edition, John Wiley & Sons, 2014. th − S. Lippman, J. Lajoie, B. E. Moo, C++ Primer, 5 edition, Addison Wesley, , 2013. th − S. Prata, C++ Primer Plus, 6 edition, Addison Wesley, 2012. − N. Josuttis, The C++ standard library. a tutorial and reference. Pearson Education. 2012. − A. Williams, C++ Concurrency in Action:Practical Multithreading. Greenwich, CT: Manning. 2012. Module 1 Introduction to C++ Introduction to C++ Content − History and evolution − Overview of the key features ● New built-in types ● Scope and namespaces ● Enumerations ● Dynamic memory: new and delete ● Smart pointers: unique_ptr, shared_ptr, weak_ptr ● Error handling with exceptions ● References ● The const modifier Introduction to C++ History and evolution − Creator: Bjarne Stroustrup 1983 − Standards: ● The first C++ standard − 1998 (C++98, major) − 2003 (C++03, minor) ● The second C++ standard − 2011 (C++11, major) – significant improvements in language and library − 2014 (C++14, minor) − 2017 (C++17, major) Introduction to C+ History and evolution − source: https://isocpp.org/std/status Introduction to C+ History and evolution − source:
    [Show full text]
  • Name, Scope and Binding (2) in Text: Chapter 5
    Name, Scope and Binding (2) In Text: Chapter 5 N. Meng, F. Poursardar Variable • A program variable is an abstraction of a memory cell or a collection of cells • It has several attributes – Name: A mnemonic character string – Address – Type 2 Variable Attributes (continued) • Storage Bindings – Allocation o Getting a memory cell from a pool of available memory to bind to a variable – Deallocation o Putting a memory cell that has been unbound from a variable back into the pool • Lifetime – The lifetime of a variable is the time during which it is bound to a particular memory cell 3 Object Lifetime and Storage Management • Key events: creation of objects, creation of bindings, references to variables (which use bindings), (temporary) deactivation of bindings, reactivation of bindings, destruction of bindings, and destruction of objects. • Binding lifetime: the period of time from creation to destruction of a name-to-object binding. • Object lifetime: the time between the creation and destruction of an objects is the object’s lifetime: – If object outlives binding it's garbage. – If binding outlives object it's a dangling reference. • Scope: the textual region of the program in which the binding is active; we sometimes use the word scope as a noun all by itself, without an indirect object. Lifetime • If an object’s memory binding outlives its access binding, we get garbage • If an object’s access binding outlives its memory binding, we get a dangling reference • Variable lifetime begins at allocation, and ends at deallocation either by the program or garbage collector 5 Categories of Variables by Lifetimes • Static • Stack-dynamic • Explicit heap-dynamic • Implicit heap-dynamic Storage Allocation Mechanisms • Static: objects are given an absolute address that is retained throughout the program’s execution.
    [Show full text]
  • Memory Management Programming Guide
    Memory Management Programming Guide Performance 2010-06-24 PROVIDED “AS IS,” AND YOU, THE READER, ARE ASSUMING THE ENTIRE RISK AS TO ITS QUALITY Apple Inc. AND ACCURACY. © 2010 Apple Inc. IN NO EVENT WILL APPLE BE LIABLE FOR DIRECT, All rights reserved. INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES RESULTING FROM ANY DEFECT OR INACCURACY IN THIS DOCUMENT, even No part of this publication may be reproduced, if advised of the possibility of such damages. stored in a retrieval system, or transmitted, in THE WARRANTY AND REMEDIES SET FORTH ABOVE any form or by any means, mechanical, ARE EXCLUSIVE AND IN LIEU OF ALL OTHERS, ORAL OR WRITTEN, EXPRESS OR IMPLIED. No Apple electronic, photocopying, recording, or dealer, agent, or employee is authorized to make otherwise, without prior written permission of any modification, extension, or addition to this Apple Inc., with the following exceptions: Any warranty. person is hereby authorized to store Some states do not allow the exclusion or limitation of implied warranties or liability for incidental or documentation on a single computer for consequential damages, so the above limitation or personal use only and to print copies of exclusion may not apply to you. This warranty gives you specific legal rights, and you may also have documentation for personal use provided that other rights which vary from state to state. the documentation contains Apple’s copyright notice. The Apple logo is a trademark of Apple Inc. Use of the “keyboard” Apple logo (Option-Shift-K) for commercial purposes without the prior written consent of Apple may constitute trademark infringement and unfair competition in violation of federal and state laws.
    [Show full text]
  • Contents Credits & Contacts
    Overload issue 63 october 2004 contents credits & contacts Overload Editor: The Encapsulate Context Pattern Alan Griffiths Allan Kelly 6 [email protected] [email protected] Microsoft Visual C++ and Win32 Contributing Editor: Structured Exception Handling Mark Radford [email protected] Roger Orr 15 A Mini-project to Decode a Advisors: Mini-language Thomas Guest 20 Phil Bass [email protected] Garbage Collection and Object Lifetime Thaddaeus Frogley Ric Parkin 24 [email protected] Richard Blundell C++ Lookup Mysteries [email protected] Sven Rosvall 28 Advertising: Chris Lowe [email protected] Overload is a publication of the ACCU. For details of the ACCU and other ACCU publications and activities, see the ACCU website. ACCU Website: http://www.accu.org/ Information and Membership: Join on the website or contact David Hodge [email protected] Publications Officer: John Merrells [email protected] ACCU Chair: Ewan Milne [email protected] 3 Overload issue 63 october 2004 Editorial: The Buzzword Adoption Pattern? n my last editorial (in Overload 60, called “An Industry That Refuses to Learn”) I asserted that the software development industry has not made significant progress Iin the last quarter of a century. This assertion provoked enough of a response to fill the letters page in the following issue. I’m pleased about that, but at the same time, not so pleased. I’m pleased because I managed to provoke people into putting pen to paper – or rather, in this day and age, putting fingers to keyboard. I’m not so pleased because the response was one of overwhelming agreement, which is unfortunate because it suggests that any hopes I may have had that my experience is the odd one out, are false.
    [Show full text]
  • Objective-C Fundamentals
    Christopher K. Fairbairn Johannes Fahrenkrug Collin Ruffenach MANNING Objective-C Fundamentals Download from Wow! eBook <www.wowebook.com> Download from Wow! eBook <www.wowebook.com> Objective-C Fundamentals CHRISTOPHER K. FAIRBAIRN JOHANNES FAHRENKRUG COLLIN RUFFENACH MANNING SHELTER ISLAND Download from Wow! eBook <www.wowebook.com> For online information and ordering of this and other Manning books, please visit www.manning.com. The publisher offers discounts on this book when ordered in quantity. For more information, please contact Special Sales Department Manning Publications Co. 20 Baldwin Road PO Box 261 Shelter Island, NY 11964 Email: [email protected] ©2012 by Manning Publications Co. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by means electronic, mechanical, photocopying, or otherwise, without prior written permission of the publisher. Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks. Where those designations appear in the book, and Manning Publications was aware of a trademark claim, the designations have been printed in initial caps or all caps. Recognizing the importance of preserving what has been written, it is Manning’s policy to have the books we publish printed on acid-free paper, and we exert our best efforts to that end. Recognizing also our responsibility to conserve the resources of our planet, Manning books are printed on paper that is at least 15 percent recycled
    [Show full text]
  • Object Lifetime Prediction in Java
    Object Lifetime Prediction in Java Hajime Inoue Darko Stefanovic´ Stephanie Forrest Department of Computer Science University of New Mexico Albuquerque, NM 87106 hinoue, darko, forrest ¡ @cs.unm.edu Technical Report TR-CS-2003-28 May 2003 Abstract Accurately predicting the lifetimes of objects in object-oriented and functional languages is im- portant because such predictions can be used to improve memory management performance at run-time. A typical approach to this prediction problem is first to observe object lifetimes by tracing a sample program execution, then to construct a predictor function based on these ob- servations, and finally to test the predictions on reference program executions. Four quantitative measures characterize this approach: coverage, the proportion of objects in the reference program execution for which the predictor function is defined; accuracy, the fraction of predicted lifetimes that are correct; precision, the granularity of the predictions; and size, the size of the predictor itself. These four properties are not independent; for example, increased precision often leads to less coverage and accuracy. We describe a fully precise prediction method and report experimental results on its perfor- mance. By “fully precise” we mean that the granularity of predictions is equal to the smallest unit of allocation. We show that for a number of benchmark programs in the Java programming lan- guage, fully precise prediction can be achieved, together with high coverage and accuracy. Our results also show that a significant proportion of objects have a measured lifetime of zero, a fact which a dynamic compiler could use to avoid explicit allocation. The method described here is the first to combine high-precision and efficiency in a single lifetime predictor.
    [Show full text]
  • On the Prediction of Java Object Lifetimes
    IEEE TRANSACTIONS ON COMPUTERS, VOL. 0, NO. 0, MONTH 2003 1 On the Prediction of Java Object Lifetimes Hajime Inoue, Darko Stefanovic,´ Member, IEEE, and Stephanie Forrest, Member, IEEE Abstract Accurately predicting object lifetimes is important for improving memory management systems. Current garbage collectors make relatively coarse-grained predictions (e.g., “short lived” vs. “long lived”) and rely on application-independent heuristics related to the local characteristics of an allocation. This paper introduces a prediction method which is fully precise and makes its predictions based on application-specific training rather than application-independent heuristics. By “fully precise” we mean that the granularity of predictions is equal to the smallest unit of allocation. The method described here is the first to combine high precision and efficiency in a single lifetime predictor. Fully precise prediction enables us for the first time to study zero-lifetime objects. The paper reports results showing that zero-lifetime objects comprise a significant fraction of object allocations in benchmark programs for the Java programming language and that they are correlated with their allocation context (the call stack and allocation site). Beyond zero-lifetime objects, the paper reports results on predicting longer lived objects, where in some cases it is possible to predict well the lifetime of objects based on their allocation context (the call stack and allocation site). For the SPEC benchmark programs, the number of dynamically allocated objects whose call sites have accurate predictors ranges from 0.2% to 61%. This method could potentially improve the performance of garbage collectors. The paper proposes a death- ordered collector (DOC) and analyzes its implementation overheads and its best possible performance.
    [Show full text]
  • Names, Scopes and Bindings
    Chapter 3: Names, Scopes, and Bindings 長庚大學資訊工程學系 陳仁暉 助 理教授 Tel: (03) 211-8800 Ext: 5990 Email: [email protected] URL: http://www.csie.cgu.edu.tw/~jhchen © All rights reserved. No part of this publication and file may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of Professor Jenhui Chen (E-mail: [email protected]). 1 High-level programming languages High-level features relative to assembly language ‘Highness’ = degree of abstraction machine independence efficient implementation does not depend on a specific machine instruction set relatively easy goal ease of programming hard goal more aesthetics(藝術的), trial and error than science ? Programming language design find the right abstractions 2 Name Mnemonic character string to represent something else usually identifiers, e.g., bull, cow, boy, girl, etc. some languages allow names like ‘+’ & ‘:=‘ Enables programmers to refer to variables etc. using symbolic names rather than (e.g., $r0, $s0, sp, or fp, etc) low-level hardware addresses (e.g., 0x3F3D or 0x11AA) abstract their control and data structures express purpose or function instead of implementation make programs manageable control abstraction: subroutines data abstraction: classes (for example) 3 Binding & Binding Time... Binding an association between a name and the thing that is named Binding time the time at which an implementation decision is made to create a binding The time spent in implementation decision Language design time the design of specific program constructs (syntax) primitive types meaning (semantics) Language implementation time fixation of implementation constants such as numeric precision run-time memory sizes max identifier name length number and types of built-in exceptions, etc.
    [Show full text]