Changes the value associated with the specified thread-specific data key.

Syntax

tis_setspecific(
key ,
value );

Argument	Data Type	Access
key	opaque pthread_key_t	read
value	void *	read

int
tis_setspecific (
pthread_key_t key,
const void *value);

Arguments

key

Thread-specific data key that identifies the data to receive value. Must be obtained from a call to tis_key_create() .

value

New value to associate with the specified key. Once set, this value can be retrieved using the same key in a call to tis_getspecific() .

Description

This routine sets the value associated with the specified thread-specific data key. If a value is defined for the key (that is, the current value is not NULL), the new value is substituted for it. The key is obtained by a previous call to tis_key_create() .

Do not call this routine from a data destructor function. Doing so could lead to a memory leak or an infinite loop.

Return Values If an error condition occurs, this routine returns an integer indicating the type of error. Possible return values are as follows:

Return	Description
0	Successful completion.
[EINVAL]	The value specified by key is not a valid key.
[ENOMEM]	Insufficient memory to associate the value with the key.

Associated Routines

tis_getspecific()
tis_key_create()
tis_key_delete()

Used as the synchronization point for asynchronous I/O system services. This routine is for OpenVMS systems only.

Syntax

tis_sync(
efn ,
iosb );

Argument	Data Type	Access
efn	unsigned long	read
iosb	void *	read

int
tis_sync (
unsigned long efn,
void *iosb);

Arguments

efn

The event flag specified with the OpenVMS system service routine.

iosb

The IOSB specified with the OpenVMS system service routine.

Description

When you are performing thread-synchronous "wait-form" system service calls on OpenVMS such as $QIOW , $ENQW , $GETJPIW , and so on, you should use this routine and tis_io_complete() with the asynchronous form of the service (that is, without the "W") and specify the address of tis_io_complete() as the completion AST routine (the AST argument, if any, is ignored). The call must also specify an IOSB (or equivalent, such as an LKSB) and if possible a unique event flag (see lib$get_ef ). Once the library code is ready to wait for the I/O, it simply calls tis_sync() (just as if it were calling $SYNC ).

Associated Routines

tis_io_complete()

Creates a cancelation point in the calling thread.

Syntax

tis_testcancel( );

void
tis_testcancel (void);

Arguments

None

Description

This routine requests delivery of a pending cancelation request to the calling thread. Thus, this routine creates a cancelation point in the calling thread. The cancelation request is delivered only if a request is pending for the calling thread and the calling thread's cancelability state is enabled. (A thread disables delivery of cancelation requests to itself by calling tis_setcancelstate() .)

This routine, when called within very long loops, ensures that a pending cancelation request is noticed within a reasonable amount of time.

Associated Routines

tis_setcancelstate()

Unlocks the Threads Library global mutex.

Syntax

tis_unlock_global( );

int
tis_unlock_global (void);

Arguments

None

Description

This routine unlocks the global mutex. Because the global mutex is recursive, the unlock occurs when each call to tis_lock_global() has been matched by a call to this routine. For example, if your program called tis_lock_global() three times, tis_unlock_global() unlocks the global mutex when you call it the third time.

For more information about actions taken when threads are present, refer to the pthread_unlock_global_np() description.

Associated Routines

tis_lock_global()

Acquires a read-write lock for write access.

Syntax

tis_write_lock(
lock );

Argument	Data Type	Access
lock	opaque tis_rwlock_t	write

int
tis_write_lock (
tis_rwlock_t *lock);

Arguments

lock

Address of the read-write lock to be acquired for write access.

Description

This routine acquires a read-write lock for write access. This routine waits for any other active locks (for either read or write access) to be unlocked before this acquisition request is granted.

This routine returns when the specified read-write lock is acquired for write access.

Associated Routines

tis_read_lock()
tis_read_trylock()
tis_read_unlock()
tis_rwlock_destroy()
tis_rwlock_init()
tis_write_trylock()
tis_write_unlock()

Attempts to acquire a read-write lock for write access.

Syntax

tis_write_trylock(
lock );

Argument	Data Type	Access
lock	opaque tis_rwlock_t	write

int
tis_write_trylock (
tis_rwlock_t *lock);

Arguments

lock

Address of the read-write lock to be acquired for write access.

Description

This routine attempts to acquire a read-write lock for write access. The routine attempts to immediately acquire the lock. If the lock is acquired, zero (0) is returned. If the lock is held by another thread (for either read or write access), [EBUSY] is returned and the calling thread does not wait for the write-access lock to be acquired.

Note that it is a coding error to attempt to acquire the lock for write access if the lock is already held by the calling thread. (However, this routine returns [EBUSY] anyway, because no ownership error-checking takes place.)

Associated Routines

tis_read_lock()
tis_read_trylock()
tis_read_unlock()
tis_rwlock_destroy()
tis_rwlock_init()
tis_write_lock()
tis_write_unlock()

Unlocks a read-write lock that was acquired for write access.

Syntax

tis_write_unlock(
lock );

Argument	Data Type	Access
lock	opaque tis_rwlock_t	write

int
tis_write_unlock (
tis_rwlock_t *lock);

Arguments

lock

Address of the read-write lock to be unlocked.

Description

This routine unlocks a read-write lock that was acquired for write access.

Upon completion of this routine, any thread waiting to acquire the lock for read access will have those acquisitions granted. If no threads are waiting to acquire the lock for read access, then a thread waiting to acquire it for write access will have that acquisition granted.

Associated Routines

tis_read_lock()
tis_read_trylock()
tis_read_unlock()
tis_rwlock_init()
tis_rwlock_destroy()
tis_write_lock()
tis_write_trylock()

Notifies the scheduler that the current thread is willing to release its processor to other threads of the same or higher priority.

Syntax

tis_yield( );

Arguments

None

Description

When threads are not present, this routine has no effect.

This routine notifies the thread scheduler that the current thread is willing to release its processor to other threads of equivalent or greater scheduling precedence. (A thread generally will release its processor to a thread of a greater scheduling precedence without calling this routine.) If no other threads of equivalent or greater scheduling precedence are ready to execute, the thread continues.

This routine can allow knowledge of the details of an application to be used to improve its performance. If a thread does not call tis_yield() , other threads may be given the opportunity to run at arbitrary points (possibly even when the interrupted thread holds a required resource). By making strategic calls to tis_yield() , other threads can be given the opportunity to run when the resources are free. This improves performance by reducing contention for the resource.

As a general guideline, consider calling this routine after a thread has released a resource (such as a mutex) which is heavily contended for by other threads. This can be especially important if the program is running on a uniprocessor machine, or if the thread acquires and releases the resource inside a tight loop.

Use this routine carefully and sparingly, because misuse can cause unnecessary context switching that will increase overhead and actually degrade performance. For example, it is counter-productive for a thread to yield while it holds a resource that the threads to which it is yielding will need. Likewise, it is pointless to yield unless there is likely to be another thread that is ready to run.

This appendix discusses Threads Library issues specific to Tru64 UNIX systems.

A.1 Overview

The Tru64 UNIX operating system supports multiple concurrent "execution contexts" within a process. The Threads Library uses these kernel execution contexts to implement user threads. One important benefit of this is that user threads can run simultaneously on separate processors in a multiprocessor system. Review Section 3.1 for tips for ensuring that your application will work correctly with kernel threads and multiprocessing.

A.2 Building Threaded Applications

The following sections discuss points to consider when building using the Threads Library.

A.2.1 Including Threads Header Files

Include one of the Threads Library header files shown in Table A-1 in your program to use the appropriate Threads library.

Table A-1 Header Files

Header File	Interface
pthread.h	POSIX routines
tis.h	Thread-independent services routines

Do not include more than one of these header files in your module.

A.2.2 Building Multithreaded Applications from Threads Libraries

Multithreaded applications are built using shared libraries. For a description of shared libraries, see the Tru64 UNIX Programmer's Guide.

Table A-2 contains the libraries supported for multithreaded programming.

Table A-2 Tru64 UNIX Shared Libraries for Multithreaded Programs

libpthreads.so	Shared version of Threads Library "legacy" package, implementing the HP-proprietary CMA (or cma) and POSIX 1003.4a/Draft 4 ( d4 or DCEthreads) interfaces.
libpthread.so	Shared version of the POSIX threads package. Requires libexc.so and libc.so
libexc.so	Shared version of Tru64 UNIX exception support package.
libc.so	Shared version of the C language run-time library ( libc.so ).

Build a multithreaded application using shared versions of libexc , libpthread , and libc using this command:

% cc -o myprog myprog.c -pthread

If you use a compiler front-end or other (not C) language environment that does not support the -pthread compilation switch, you must provide the -D_REENTRANT compilation switch (or equivalent) at compilation, and link as shown in Section A.2.3.

A.2.3 Linking Multithreaded Shared Libraries

The ld command does not support the -pthread or -threads switch. Normally, programs can be compiled and linked from the cc command. If you must link using the ld command, you must list the shared libraries in the proper order. The libc library should be the last library referenced, libexc should immediately precede libc , and the thread libraries should precede libexc .

For libraries that use only the pthread interface, use the following:

ld <...> -lpthread -lexc -lc

If using the cma or d4 interfaces, use the following:

ld <...> -lpthreads -lpthread -lexc -lc

Also, cc -pthread (or cc -threads ) causes the compiler to replace any libraries that have special thread-safe alternatives. These libraries have the same name ending in -r . For example, cc -pthread -o foo -lbar , if there is a libbar.so and libbar_r.so , would use the latter. When linking with the ld command, you must perform that search and replacement yourself.