book

Advanced Programming in the UNIX® Environment: Second Edition

by W. Richard Stevens, Stephen A. Rago

June 2005

Intermediate to advanced

960 pages

23h 41m

English

Addison-Wesley Professional

Read now

Unlock full access

Dedication
IntroductionChanges from the First EditionAcknowledgments
IntroductionUnix StandardsOrganization of the BookExamples in the TextSystems Used to Test the ExamplesAcknowledgments
1.1. Introduction1.2. UNIX Architecture1.3. Logging InLogin NameShells1.4. Files and DirectoriesFile SystemFilenamePathnameWorking DirectoryHome Directory1.5. Input and OutputFile DescriptorsStandard Input, Standard Output, and Standard ErrorUnbuffered I/OStandard I/O1.6. Programs and ProcessesProgramProcesses and Process IDProcess ControlThreads and Thread IDs1.7. Error HandlingError Recovery1.8. User IdentificationUser IDGroup IDSupplementary Group IDs1.9. Signals1.10. Time Values1.11. System Calls and Library Functions1.12. SummaryExercises
2.1. Introduction2.2. UNIX Standardization2.2.1. ISO C2.2.2. IEEE POSIX2.2.3. The Single UNIX Specification2.2.4. FIPS2.3. UNIX System Implementations2.3.1. UNIX System V Release 42.3.2. 4.4BSD2.3.3. FreeBSD2.3.4. Linux2.3.5. Mac OS X2.3.6. Solaris2.3.7. Other UNIX Systems2.4. Relationship of Standards and Implementations2.5. Limits2.5.1. ISO C Limits2.5.2. POSIX Limits2.5.3. XSI Limits2.5.4. sysconf, pathconf, and fpathconf Functions2.5.5. Indeterminate Runtime LimitsPathnamesMaximum Number of Open Files2.6. Options2.7. Feature Test Macros2.8. Primitive System Data Types2.9. Conflicts Between Standards2.10. SummaryExercises
3.1. Introduction3.2. File Descriptors3.3. open FunctionFilename and Pathname Truncation3.4. creat Function3.5. close Function3.6. lseek Function3.7. read Function3.8. write Function3.9. I/O Efficiency3.10. File Sharing3.11. Atomic OperationsAppending to a Filepread and pwrite FunctionsCreating a File3.12. dup and dup2 Functions3.13. sync, fsync, and fdatasync Functions3.14. fcntl Function3.15. ioctl Function3.16. /dev/fd3.17. SummaryExercises

4.1. Introduction4.2. stat, fstat, and lstat Functions4.3. File Types4.4. Set-User-ID and Set-Group-ID4.5. File Access Permissions4.6. Ownership of New Files and Directories4.7. access Function4.8. umask Function4.9. chmod and fchmod Functions4.10. Sticky Bit4.11. chown, fchown, and lchown Functions4.12. File SizeHoles in a File4.13. File Truncation4.14. File Systems4.15. link, unlink, remove, and rename Functions4.16. Symbolic Links4.17. symlink and readlink Functions4.18. File Times4.19. utime Function4.20. mkdir and rmdir Functions4.21. Reading Directories4.22. chdir, fchdir, and getcwd Functions4.23. Device Special Files4.24. Summary of File Access Permission Bits4.25. SummaryExercises
5.1. Introduction5.2. Streams and FILE Objects5.3. Standard Input, Standard Output, and Standard Error5.4. Buffering5.5. Opening a Stream5.6. Reading and Writing a StreamInput FunctionsOutput Functions5.7. Line-at-a-Time I/O5.8. Standard I/O Efficiency5.9. Binary I/O5.10. Positioning a Stream5.11. Formatted I/OFormatted OutputFormatted Input5.12. Implementation Details5.13. Temporary Files5.14. Alternatives to Standard I/O5.15. SummaryExercises
6.1. Introduction6.2. Password File6.3. Shadow Passwords6.4. Group File6.5. Supplementary Group IDs6.6. Implementation Differences6.7. Other Data Files6.8. Login Accounting6.9. System Identification6.10. Time and Date Routines6.11. SummaryExercises
7.1. Introduction7.2. main Function7.3. Process TerminationExit Functionsatexit Function7.4. Command-Line Arguments7.5. Environment List7.6. Memory Layout of a C Program7.7. Shared Libraries7.8. Memory AllocationAlternate Memory Allocatorslibmallocvmallocquick-fitalloca Function7.9. Environment Variables7.10. setjmp and longjmp FunctionsAutomatic, Register, and Volatile VariablesPotential Problem with Automatic Variables7.11. getrlimit and setrlimit Functions7.12. SummaryExercises
8.1. Introduction8.2. Process Identifiers8.3. fork FunctionFile Sharing8.4. vfork Function8.5. exit Functions8.6. wait and waitpid Functions8.7. waitid Function8.8. wait3 and wait4 Functions8.9. Race Conditions8.10. exec Functions8.11. Changing User IDs and Group IDssetreuid and setregid Functionsseteuid and setegid FunctionsGroup IDs8.12. Interpreter Files8.13. system FunctionSet-User-ID Programs8.14. Process Accounting8.15. User Identification8.16. Process Times8.17. SummaryExercises
9.1. Introduction9.2. Terminal LoginsBSD Terminal LoginsMac OS X Terminal LoginsLinux Terminal LoginsSolaris Terminal Logins9.3. Network LoginsBSD Network LoginsMac OS X Network LoginsLinux Network LoginsSolaris Network Logins9.4. Process Groups9.5. Sessions9.6. Controlling Terminal9.7. tcgetpgrp, tcsetpgrp, and tcgetsid Functions9.8. Job Control9.9. Shell Execution of Programs9.10. Orphaned Process Groups9.11. FreeBSD Implementation9.12. SummaryExercises
10.1. Introduction10.2. Signal Concepts10.3. signal FunctionProgram Start-UpProcess Creation10.4. Unreliable Signals10.5. Interrupted System Calls10.6. Reentrant Functions10.7. SIGCLD Semantics10.8. Reliable-Signal Terminology and Semantics10.9. kill and raise Functions10.10. alarm and pause Functions10.11. Signal SetsImplementation10.12. sigprocmask Function10.13. sigpending Function10.14. sigaction Function10.15. sigsetjmp and siglongjmp Functions10.16. sigsuspend Function10.17. abort Function10.18. system FunctionReturn Value from system10.19. sleep Function10.20. Job-Control Signals10.21. Additional FeaturesSignal NamesSignal Mappings10.22. SummaryExercises
11.1. Introduction11.2. Thread Concepts11.3. Thread Identification11.4. Thread Creation11.5. Thread Termination11.6. Thread SynchronizationMutexesDeadlock AvoidanceReader–Writer LocksCondition Variables11.7. SummaryExercises
12.1. Introduction12.2. Thread Limits12.3. Thread AttributesMore Thread Attributes12.4. Synchronization AttributesMutex AttributesReader–Writer Lock AttributesCondition Variable Attributes12.5. Reentrancy12.6. Thread-Specific Data12.7. Cancel Options12.8. Threads and Signals12.9. Threads and fork12.10. Threads and I/O12.11. SummaryExercises
13.1. Introduction13.2. Daemon Characteristics13.3. Coding Rules13.4. Error Logging13.5. Single-Instance Daemons13.6. Daemon Conventions13.7. Client–Server Model13.8. SummaryExercises
14.1. Introduction14.2. Nonblocking I/O14.3. Record LockingHistoryfcntl Record LockingImplied Inheritance and Release of LocksFreeBSD ImplementationLocks at End of FileAdvisory versus Mandatory Locking14.4. STREAMSSTREAMS Messagesputmsg and putpmsg FunctionsSTREAMS ioctl Operationswrite to STREAMS DevicesWrite Modegetmsg and getpmsg FunctionsRead Mode14.5. I/O Multiplexing14.5.1. select and pselect Functions14.5.2. poll FunctionInterruptibility of select and poll14.6. Asynchronous I/O14.6.1. System V Asynchronous I/O14.6.2. BSD Asynchronous I/O14.7. readv and writev Functions14.8. readn and writen Functions14.9. Memory-Mapped I/O14.10. SummaryExercises
15.1. Introduction15.2. Pipes15.3. popen and pclose Functions15.4. Coprocesses15.5. FIFOs15.6. XSI IPC15.6.1. Identifiers and Keys15.6.2. Permission Structure15.6.3. Configuration Limits15.6.4. Advantages and Disadvantages15.7. Message Queues15.8. SemaphoresSemaphore Adjustment on exit15.9. Shared Memory15.10. Client–Server Properties15.11. SummaryExercises
16.1. Introduction16.2. Socket Descriptors16.3. Addressing16.3.1. Byte Ordering16.3.2. Address Formats16.3.3. Address Lookup16.3.4. Associating Addresses with Sockets16.4. Connection Establishment16.5. Data Transfer16.6. Socket Options16.7. Out-of-Band Data16.8. Nonblocking and Asynchronous I/O16.9. SummaryExercises
17.1. Introduction17.2. STREAMS-Based Pipes17.2.1. Naming STREAMS Pipes17.2.2. Unique Connections17.3. UNIX Domain Sockets17.3.1. Naming UNIX Domain Sockets17.3.2. Unique Connections17.4. Passing File Descriptors17.4.1. Passing File Descriptors over STREAMS-Based Pipes17.4.2. Passing File Descriptors over UNIX Domain Sockets17.5. An Open Server, Version 117.6. An Open Server, Version 217.7. SummaryExercises
18.1. Introduction18.2. Overview18.3. Special Input Characters18.4. Getting and Setting Terminal Attributes18.5. Terminal Option Flags18.6. stty Command18.7. Baud Rate Functions18.8. Line Control Functions18.9. Terminal Identification18.10. Canonical Mode18.11. Noncanonical Mode18.12. Terminal Window Size18.13. termcap, terminfo, and curses18.14. SummaryExercises
19.1. Introduction19.2. OverviewNetwork Login Serversscript Programexpect ProgramRunning CoprocessesWatching the Output of Long-Running Programs19.3. Opening Pseudo-Terminal Devices19.3.1. STREAMS-Based Pseudo Terminals19.3.2. BSD-Based Pseudo Terminals19.3.3. Linux-Based Pseudo Terminals19.4. pty_fork Function19.5. pty Program19.6. Using the pty Programutmp FileJob Control InteractionWatching the Output of Long-Running Programsscript ProgramRunning CoprocessesDriving Interactive Programs Noninteractively19.7. Advanced FeaturesPacket ModeRemote ModeWindow Size ChangesSignal Generation19.8. SummaryExercises
20.1. Introduction20.2. History20.3. The Library20.4. Implementation Overview20.5. Centralized or Decentralized?20.6. ConcurrencyCoarse-Grained LockingFine-Grained Locking20.7. Building the Library20.8. Source Code20.9. PerformanceSingle-Process ResultsMultiple-Process Results20.10. SummaryExercises
21.1. Introduction21.2. The Internet Printing Protocol21.3. The Hypertext Transfer Protocol21.4. Printer SpoolingSecurity21.5. Source Code21.6. SummaryExercises
B.1. Our Header FileB.2. Standard Error Routines
Chapter 1Chapter 2Chapter 3Chapter 4Chapter 5Chapter 6Chapter 7Chapter 8Chapter 9Chapter 10Chapter 11Chapter 12Chapter 13Chapter 14Chapter 15Chapter 16Chapter 17Chapter 18Chapter 19Chapter 20Chapter 21

Content preview from Advanced Programming in the UNIX® Environment: Second Edition

Chapter 10. Signals

Introduction

Signals are software interrupts. Most nontrivial application programs need to deal with signals. Signals provide a way of handling asynchronous events: a user at a terminal typing the interrupt key to stop a program or the next program in a pipeline terminating prematurely.

Signals have been provided since the early versions of the UNIX System, but the signal model provided with systems such as Version 7 was not reliable. Signals could get lost, and it was difficult for a process to turn off selected signals when executing critical regions of code. Both 4.3BSD and SVR3 made changes to the signal model, adding what are called reliable signals. But the changes made by Berkeley and AT&T were incompatible. Fortunately, ...