本文转自：



POSIX Threads Programming
Blaise Barney, Lawrence Livermore National Laboratory	UCRL-MI-133316

Table of Contents

Abstract


In shared memory multiprocessor architectures, such as SMPs, threads can be used to implement parallelism. Historically, hardware vendors have implemented their own proprietary versions of threads, making portability a concern for software developers. For UNIX systems, a standardized C language threads programming interface has been specified by the IEEE POSIX 1003.1c standard. Implementations that adhere to this standard are referred to as POSIX threads, or Pthreads.

The tutorial begins with an introduction to concepts, motivations, and design considerations for using Pthreads. Each of the three major classes of routines in the Pthreads API are then covered: Thread Management, Mutex Variables, and Condition Variables. Example codes are used throughout to demonstrate how to use most of the Pthreads routines needed by a new Pthreads programmer. The tutorial concludes with a discussion of LLNL specifics and how to mix MPI with pthreads. A lab exercise, with numerous example codes (C Language) is also included.

Level/Prerequisites: Ideal for those who are new to parallel programming with threads. A basic understanding of parallel programming in C is assumed. For those who are unfamiliar with Parallel Programming in general, the material covered in  would be helpful. 



Pthreads Overview

What is a Thread?

• Technically, a thread is defined as an independent stream of instructions that can be scheduled to run as such by the operating system. But what does this mean?

• To the software developer, the concept of a "procedure" that runs independently from its main program may best describe a thread.

• To go one step further, imagine a main program (a.out) that contains a number of procedures. Then imagine all of these procedures being able to be scheduled to run simultaneously and/or independently by the operating system. That would describe a "multi-threaded" program.

• How is this accomplished?
• Before understanding a thread, one first needs to understand a UNIX process. A process is created by the operating system, and requires a fair amount of "overhead". Processes contain information about program resources and program execution state, including:
  • Process ID, process group ID, user ID, and group ID
  • Environment
  • Working directory.
  • Program instructions
  • Registers
  • Stack
  • Heap
  • File descriptors
  • Signal actions
  • Shared libraries
  • Inter-process communication tools (such as message queues, pipes, semaphores, or shared memory).

  UNIX PROCESS	THREADS WITHIN A UNIX PROCESS

• Threads use and exist within these process resources, yet are able to be scheduled by the operating system and run as independent entities largely because they duplicate only the bare essential resources that enable them to exist as executable code.

• This independent flow of control is accomplished because a thread maintains its own:
  • Stack pointer
  • Registers
  • Scheduling properties (such as policy or priority)
  • Set of pending and blocked signals
  • Thread specific data.

• So, in summary, in the UNIX environment a thread:
  • Exists within a process and uses the process resources
  • Has its own independent flow of control as long as its parent process exists and the OS supports it
  • Duplicates only the essential resources it needs to be independently schedulable
  • May share the process resources with other threads that act equally independently (and dependently)
  • Dies if the parent process dies - or something similar
  • Is "lightweight" because most of the overhead has already been accomplished through the creation of its process.

• Because threads within the same process share resources:
  • Changes made by one thread to shared system resources (such as closing a file) will be seen by all other threads.
  • Two pointers having the same value point to the same data.
  • Reading and writing to the same memory locations is possible, and therefore requires explicit synchronization by the programmer.

Pthreads Overview

What are Pthreads?

• Historically, hardware vendors have implemented their own proprietary versions of threads. These implementations differed substantially from each other making it difficult for programmers to develop portable threaded applications.

• In order to take full advantage of the capabilities provided by threads, a standardized programming interface was required.
  • For UNIX systems, this interface has been specified by the IEEE POSIX 1003.1c standard (1995).
  • Implementations adhering to this standard are referred to as POSIX threads, or Pthreads.
  • Most hardware vendors now offer Pthreads in addition to their proprietary API's.

• The POSIX standard has continued to evolve and undergo revisions, including the Pthreads specification. The latest version is known as IEEE Std 1003.1, 2004 Edition.

• Some useful links:
  • POSIX FAQs: 
  • Download the Standard: 

• Pthreads are defined as a set of C language programming types and procedure calls, implemented with a pthread.h header/include file and a thread library - though this library may be part of another library, such as libc, in some implementations.

Pthreads Overview

Why Pthreads?

• The primary motivation for using Pthreads is to realize potential program performance gains.

• When compared to the cost of creating and managing a process, a thread can be created with much less operating system overhead. Managing threads requires fewer system resources than managing processes.

  For example, the following table compares timing results for the fork() subroutine and the pthread_create() subroutine. Timings reflect 50,000 process/thread creations, were performed with the time utility, and units are in seconds, no optimization flags.

  Note: don't expect the sytem and user times to add up to real time, because these are SMP systems with multiple CPUs working on the problem at the same time. At best, these are approximations run on local machines, past and present.

  Platform	fork()			pthread_create()
  	real	user	sys	real	user	sys
  AMD 2.3 GHz Opteron (16cpus/node)	12.5	1.0	12.5	1.2	0.2	1.3
  AMD 2.4 GHz Opteron (8cpus/node)	17.6	2.2	15.7	1.4	0.3	1.3
  IBM 4.0 GHz POWER6 (8cpus/node)	9.5	0.6	8.8	1.6	0.1	0.4
  IBM 1.9 GHz POWER5 p5-575 (8cpus/node)	64.2	30.7	27.6	1.7	0.6	1.1
  IBM 1.5 GHz POWER4 (8cpus/node)	104.5	48.6	47.2	2.1	1.0	1.5
  INTEL 2.4 GHz Xeon (2 cpus/node)	54.9	1.5	20.8	1.6	0.7	0.9
  INTEL 1.4 GHz Itanium2 (4 cpus/node)	54.5	1.1	22.2	2.0	1.2	0.6

   fork_vs_thread.txt

• All threads within a process share the same address space. Inter-thread communication is more efficient and in many cases, easier to use than inter-process communication.

• Threaded applications offer potential performance gains and practical advantages over non-threaded applications in several other ways:
  • Overlapping CPU work with I/O: For example, a program may have sections where it is performing a long I/O operation. While one thread is waiting for an I/O system call to complete, CPU intensive work can be performed by other threads.
  • Priority/real-time scheduling: tasks which are more important can be scheduled to supersede or interrupt lower priority tasks.
  • Asynchronous event handling: tasks which service events of indeterminate frequency and duration can be interleaved. For example, a web server can both transfer data from previous requests and manage the arrival of new requests.

• The primary motivation for considering the use of Pthreads on an SMP architecture is to achieve optimum performance. In particular, if an application is using MPI for on-node communications, there is a potential that performance could be greatly improved by using Pthreads for on-node data transfer instead.

• For example:
  • MPI libraries usually implement on-node task communication via shared memory, which involves at least one memory copy operation (process to process).
  • For Pthreads there is no intermediate memory copy required because threads share the same address space within a single process. There is no data transfer, per se. It becomes more of a cache-to-CPU or memory-to-CPU bandwidth (worst case) situation. These speeds are much higher.
  • Some local comparisons are shown below:

    Platform	MPI Shared Memory Bandwidth (GB/sec)	Pthreads Worst Case Memory-to-CPU Bandwidth  (GB/sec)
    AMD 2.3 GHz Opteron	1.8	5.3
    AMD 2.4 GHz Opteron	1.2	5.3
    IBM 1.9 GHz POWER5 p5-575	4.1	16
    IBM 1.5 GHz POWER4	2.1	4
    Intel 2.4 GHz Xeon	0.3	4.3
    Intel 1.4 GHz Itanium 2	1.8	6.4

Pthreads Overview

Designing Threaded Programs

 Parallel Programming:
• On modern, multi-cpu machines, pthreads are ideally suited for parallel programming, and whatever applies to parallel programming in general, applies to parallel pthreads programs.

• There are many considerations for designing parallel programs, such as:
  • What type of parallel programming model to use?
  • Problem partitioning
  • Load balancing
  • Communications
  • Data dependencies
  • Synchronization and race conditions
  • Memory issues
  • I/O issues
  • Program complexity
  • Programmer effort/costs/time
  • ...

• Covering these topics is beyond the scope of this tutorial, however interested readers can obtain a quick overview in the  tutorial.

• In general though, in order for a program to take advantage of Pthreads, it must be able to be organized into discrete, independent tasks which can execute concurrently. For example, if routine1 and routine2 can be interchanged, interleaved and/or overlapped in real time, they are candidates for threading.
  

• Programs having the following characteristics may be well suited for pthreads:
  • Work that can be executed, or data that can be operated on, by multiple tasks simultaneously
  • Block for potentially long I/O waits
  • Use many CPU cycles in some places but not others
  • Must respond to asynchronous events
  • Some work is more important than other work (priority interrupts)

• Pthreads can also be used for serial applications, to emulate parallel execution. A perfect example is the typical web browser, which for most people, runs on a single cpu desktop/laptop machine. Many things can "appear" to be happening at the same time.

• Several common models for threaded programs exist:

  • Manager/worker: a single thread, the manager assigns work to other threads, the workers. Typically, the manager handles all input and parcels out work to the other tasks. At least two forms of the manager/worker model are common: static worker pool and dynamic worker pool.

  • Pipeline: a task is broken into a series of suboperations, each of which is handled in series, but concurrently, by a different thread. An automobile assembly line best describes this model.

  • Peer: similar to the manager/worker model, but after the main thread creates other threads, it participates in the work.

 Shared Memory Model:

• All threads have access to the same global, shared memory

• Threads also have their own private data

• Programmers are responsible for synchronizing access (protecting) globally shared data.

 Thread-safeness:

• Thread-safeness: in a nutshell, refers an application's ability to execute multiple threads simultaneously without "clobbering" shared data or creating "race" conditions.

• For example, suppose that your application creates several threads, each of which makes a call to the same library routine:
  • This library routine accesses/modifies a global structure or location in memory.
  • As each thread calls this routine it is possible that they may try to modify this global structure/memory location at the same time.
  • If the routine does not employ some sort of synchronization constructs to prevent data corruption, then it is not thread-safe.

• The implication to users of external library routines is that if you aren't 100% certain the routine is thread-safe, then you take your chances with problems that could arise.

• Recommendation: Be careful if your application uses libraries or other objects that don't explicitly guarantee thread-safeness. When in doubt, assume that they are not thread-safe until proven otherwise. This can be done by "serializing" the calls to the uncertain routine, etc.

The Pthreads API




• The original Pthreads API was defined in the ANSI/IEEE POSIX 1003.1 - 1995 standard. The POSIX standard has continued to evolve and undergo revisions, including the Pthreads specification. The latest version is known as IEEE Std 1003.1, 2004 Edition.

• Copies of the standard can be purchased from IEEE or downloaded for free from .

• The subroutines which comprise the Pthreads API can be informally grouped into four major groups:

  1. Thread management: Routines that work directly on threads - creating, detaching, joining, etc. They also include functions to set/query thread attributes (joinable, scheduling etc.)

  2. Mutexes: Routines that deal with synchronization, called a "mutex", which is an abbreviation for "mutual exclusion". Mutex functions provide for creating, destroying, locking and unlocking mutexes. These are supplemented by mutex attribute functions that set or modify attributes associated with mutexes.

  3. Condition variables: Routines that address communications between threads that share a mutex. Based upon programmer specified conditions. This group includes functions to create, destroy, wait and signal based upon specified variable values. Functions to set/query condition variable attributes are also included.

  4. Synchronization: Routines that manage read/write locks and barriers.

• Naming conventions: All identifiers in the threads library begin with pthread_. Some examples are shown below.

  Routine Prefix	Functional Group
  pthread_	Threads themselves and miscellaneous subroutines
  pthread_attr_	Thread attributes objects
  pthread_mutex_	Mutexes
  pthread_mutexattr_	Mutex attributes objects.
  pthread_cond_	Condition variables
  pthread_condattr_	Condition attributes objects
  pthread_key_	Thread-specific data keys
  pthread_rwlock_	Read/write locks
  pthread_barrier_	Synchronization barriers

• The concept of opaque objects pervades the design of the API. The basic calls work to create or modify opaque objects - the opaque objects can be modified by calls to attribute functions, which deal with opaque attributes.

• The Pthreads API contains around 100 subroutines. This tutorial will focus on a subset of these - specifically, those which are most likely to be immediately useful to the beginning Pthreads programmer.

• For portability, the pthread.h header file should be included in each source file using the Pthreads library.

• The current POSIX standard is defined only for the C language. Fortran programmers can use wrappers around C function calls. Some Fortran compilers (like IBM AIX Fortran) may provide a Fortram pthreads API.

• A number of excellent books about Pthreads are available. Several of these are listed in the section of this tutorial.

Compiling Threaded Programs




• Several examples of compile commands used for pthreads codes are listed in the table below.

  Compiler / Platform	Compiler Command	Description
  IBM  AIX	xlc_r  /  cc_r	C (ANSI  /  non-ANSI)
  	xlC_r	C++
  	xlf_r -qnosave xlf90_r -qnosave	Fortran - using IBM's Pthreads API (non-portable)
  INTEL Linux	icc -pthread	C
  	icpc -pthread	C++
  PathScale Linux	pathcc -pthread	C
  	pathCC -pthread	C++
  PGI Linux	pgcc -lpthread	C
  	pgCC -lpthread	C++
  GNU Linux, AIX	gcc -pthread	GNU C
  	g++ -pthread	GNU C++