Sponsored Content
Full Discussion: Shared static library
Top Forums UNIX for Dummies Questions & Answers Shared static library Post 302612825 by jim mcnamara on Monday 26th of March 2012 11:11:31 AM
Old 03-26-2012
There are static libraries - files usually in the /lib directory that end with .a.
There are shared libraries - files usually in the /lib directory that end with .so (or .sl), sometimes called dynamic libraries.

To link means 'Resolve external symbols or functions in your code like printf()'

By default, when you compile C code, you liink against a set of shared libraries.
You have to go out of your way to create an executable that is linked against a static library. Read your compiler documentation on how to do this.

There are a few valid uses for linking statically. Linking against shared libraries, also called dynamic linking, creates an executable that makes much better use of system resources, because eveybody else's executable file shares the same library with your code. MUCH less memory usage. That is why dynamic linking is better.

Got all that?

Next: LD_PRELOAD

This is an envrionment variable that anyone can set. If changes how the UNIX system looks for dynamic libraries. When you first run an executable, you have to find and open the dynamic libraries you linked against so all of the symbols like printf() will be there to use.

You use LD_PRELOAD to force the system to kind of auto-magically link against a library that was NOT one of the original linked libraries. So now the printf() function calls in your executable employ a different library with different code to execute printf.

There are limitations on LD_PRELOAD, to prevent bad guys from doing bad things, but that is beyond the scope of your question.
This User Gave Thanks to jim mcnamara For This Post:
 

10 More Discussions You Might Find Interesting

1. Programming

Shared memory in shared library

I need to create a shared library to access an in memory DB. The DB is not huge, but big enough to make it cumbersome to carry around in every single process using the shared library. Luckily, it is pretty static information, so I don't need to worry much about synchronizing the data between... (12 Replies)
Discussion started by: DreamWarrior
12 Replies

2. UNIX for Advanced & Expert Users

static and shared libraries

can someone explain whether my understanding is correct lets suppose we have a program that uses library x. if x is static then the code of x will be part of our program, so if we're going to have 5 executables of our program, then each executable will have x as part of it. Also, x does not... (2 Replies)
Discussion started by: JamesByars
2 Replies

3. Programming

Shared memory for shared library

I am writing a shared library in Linux (but compatible with other UNIXes) and I want to allow multiple instances to share a piece of memory -- 1 byte is enough. What's the "best" way to do this? I want to optimize for speed and portability. Obviously, I'll have to worry about mutual exclusion. (0 Replies)
Discussion started by: otheus
0 Replies

4. Shell Programming and Scripting

How to change a Makefile from building static library to shared library?

Hi: I have a library that it only offers Makefile for building static library. It built libxxx.a file. How do I in any way build a shared library? (either changin the Makefile or direct script or command to build shared library) Thanks. (1 Reply)
Discussion started by: cpthk
1 Replies

5. Linux

Could static library include static library?

I have some static library(libxxx.a libyyy.a). And I want to generate my library(libzzz.a), libzzz.a will use libxxx.a and libyyy.a I wan't my application only use libzzz.a, (means libzzz.a had include libxxx.a, libyyy.a), how can I do that? Thank you. example: I have zzz.c. I do ... (4 Replies)
Discussion started by: freemagic
4 Replies

6. Programming

Static and Shared Library in Makefile

I am having a devil of a time with a very simple make file. The program needs two shared and one static library. If I link the shared libraries only like below the mysql test app works ... (1 Reply)
Discussion started by: jadsys
1 Replies

7. Programming

Shared, Static , Dynamic?

if I could compile the same source file as shared/static/dynamic what are the advantages/ disadv of each. PS:by dynamic i am asking about usage of "dlopen". How is it particularly diff from shared libs (2 Replies)
Discussion started by: dragonpoint
2 Replies

8. Programming

Even the Static cURL Library Isn't Static

I'm writing a program which uses curl to be run on Linux PCs which will be used by a number of different users. I cannot make the users all install curl on their individual machines, so I have tried to link curl in statically, rather than using libcurl.so. I downloaded the source and created a... (8 Replies)
Discussion started by: BrandonShw
8 Replies

9. Programming

Shared library with acces to shared memory.

Hello. I am new to this forum and I would like to ask for advice about low level POSIX programming. I have to implement a POSIX compliant C shared library. A file will have some variables and the shared library will have some functions which need those variables. There is one special... (5 Replies)
Discussion started by: iamjag
5 Replies

10. AIX

Add shared members from library to same library in a different directory

I'm trying to install libiconv to AIX 7.1 from an rpm off of the perzl site. The rpm appears to install but I get this error message. add shr4.o shared members from /usr/lib/libiconv.a to /opt/freeware/lib/libiconv.a add shr.o shared members from /usr/lib/libiconv.a to ... (5 Replies)
Discussion started by: kneemoe
5 Replies
pthread_stubs(5)						File Formats Manual						  pthread_stubs(5)

NAME
pthread_stubs - list of pthread calls for which the stubs are provided in the C library DESCRIPTION
The libc shared libraries in libc cumulative patches, and onwards, contain stubs for the pthread functions in and The stubs allow non- threaded applications to dynamically load thread-safe libraries successfully, so that the pthread symbols are resolved. Applications that resolve pthread/cma calls to the stub must be built without or on the link line. Stubs provided in do not have any functionality, these are dummy functions returning zero, except the pthread_getspecific(3T) family of APIs, which have full functionality implemented in the stubs. The pthread calls to any of the stub functions below return zero. pthread_atfork(3T) pthread_attr_destroy(3T) pthread_attr_getdetachstate(3T) pthread_attr_getguardsize(3T) pthread_attr_getinheritsched(3T) pthread_attr_getschedparam(3T) pthread_attr_getschedpolicy(3T) pthread_attr_getscope(3T) pthread_attr_getstackaddr(3T) pthread_attr_getstacksize(3T) pthread_attr_setdetachstate(3T) pthread_attr_setguardsize(3T) pthread_attr_setinheritsched(3T) pthread_attr_setschedparam(3T) pthread_attr_setschedpolicy(3T) pthread_attr_setscope(3T) pthread_attr_setstackaddr(3T) pthread_attr_setstacksize(3T) pthread_cancel(3T) pthread_cond_broadcast(3T) pthread_cond_destroy(3T) pthread_cond_init(3T) pthread_cond_signal(3T) pthread_cond_timedwait(3T) pthread_cond_wait(3T) pthread_condattr_destroy(3T) pthread_condattr_getpshared(3T) pthread_condattr_init(3T) pthread_condattr_setpshared(3T) pthread_continue(3T) pthread_detach(3T) pthread_getconcurrency(3T) pthread_getschedparam(3T) pthread_join(3T) pthread_kill(3T) pthread_mutex_destroy(3T) pthread_mutex_getprioceiling(3T) pthread_mutex_init(3T) pthread_mutex_lock(3T) pthread_mutex_setprioceiling(3T) pthread_mutex_trylock(3T) pthread_mutex_unlock(3T) pthread_mutexattr_destroy(3T) pthread_mutexattr_getprioceiling(3T) pthread_mutexattr_getprotocol(3T) pthread_mutexattr_getpshared(3T) pthread_mutexattr_gettype(3T) pthread_mutexattr_init(3T) pthread_mutexattr_setprioceiling(3T) pthread_mutexattr_setprotocol(3T) pthread_mutexattr_setpshared(3T) pthread_mutexattr_settype(3T) pthread_once(3T) pthread_rwlock_destroy(3T) pthread_rwlock_init(3T) pthread_rwlock_rdlock(3T) pthread_rwlock_tryrdlock(3T) pthread_rwlock_trywrlock(3T) pthread_rwlock_unlock(3T) pthread_rwlock_wrlock(3T) pthread_rwlockattr_destroy(3T) pthread_rwlockattr_getpshared(3T) pthread_rwlockattr_init(3T) pthread_rwlockattr_setpshared(3T) pthread_self(3T) pthread_setcancelstate(3T) pthread_setcanceltype(3T) pthread_setconcurrency(3T) pthread_setschedparam(3T) pthread_sigmask(3T) pthread_suspend(3T) pthread_testcancel(3T) The stubs for the following pthread calls have full functionality. Refer to pthread(3T) for more details. pthread_key_create(3T) pthread_getspecific(3T) pthread_setspecific(3T) pthread_key_delete(3T) pthread_exit(3T) Calls to the stubs listed below, pthread_self(3T) always returns 1. returns (arg1==arg2). pthread_create(3T) and pthread_attr_init(3T) return The above mentioned stubs are provided in libc because on HP-UX if a non-threaded application links to a thread-safe library, calls to thread-safe routines from the application fail at run time due to unresolved symbols of the form of To resolve these symbols it is neces- sary to link the non-threaded application to a threads library or However, linking to a threads library forces the application to use thread-safe features even if it creates no threads, resulting in a subsequent loss of performance. To overcome the above problem, stubs for APIs have been provided in the C library. Providing stubs for API's in the HP-UX C language library have two direct effects for non-threaded applications: o thread symbols are resolved if a non-threaded application links to a thread-safe library. o Avoids the overhead of a real thread library. Especially the overhead associated with mutexes when the non-threaded application uses thread stubs rather than the real thread library procedures. Link Order Problems An application may inadvertently pick up the stubs present in when it intended to use the real pthread APIs, or cma APIs, due to link order issues. An application that needs cma behavior must link to and must do so in the supported link order, i.e. the link line should only be shared and should not contain before As long as this condition is met, the correct cma functions will be referenced. Similarly, a multi- threaded application that needs pthread library behavior must link to libpthread and must do so in a supported link order, and only use shared and EXAMPLES
Below are examples of potential link order problems. Example 1 An application or any library linked, that requires pthread/cma calls to resolve to the pthread stubs in must be built without or on the link line. If is specified before or on the link line, pthread/cma calls resolve to pthread stubs in This may lead to problems as given in the exam- ples below: $ cat thread.c #include <pthread.h> #include <stdio.h> void *thread_nothing(void *p) { printf("Success "); } int main() { int err; pthread_t thrid; err = pthread_create(&thrid, (pthread_attr_t *) NULL, thread_nothing, (void *) NULL); sleep(1); if (err) { printf("Error "); return err; } } $ cc thread.c -lc -lpthread $ a.out Error $ chatr a.out a.out: shared executable shared library dynamic path search: SHLIB_PATH disabled second embedded path disabled first Not Defined shared library list: dynamic /usr/lib/libc.2 <- libc before libpthread dynamic /usr/lib/libpthread.1 shared library binding: deferred global hash table disabled ... Solution for Example 1 For threaded applications, run the executable with environment variable set to the library or link the executable with $ LD_PRELOAD="/usr/lib/libpthread.1" a.out Success $ cc thread.c -lpthread $ a.out Success $ chatr a.out a.out: shared executable shared library dynamic path search: SHLIB_PATH disabled second embedded path disabled first Not Defined shared library list: dynamic /usr/lib/libpthread.1 dynamic /usr/lib/libc.2 shared library binding: deferred global hash table disabled ... Example 2 Specifying before in threaded applications can cause run-time problems like the following because the pthread calls get resolved to stubs in rather than the functions in pthread library. o Calls to pthread functions fail, due to uninitialized internal structures. o Calls to gethostbyname(3N) fail and return null. o Apache webmin and perl DBI applications fail with the following error message: o Calls to shl_load(3X) fail with the following error: because the stub returns zero. $ cat a.c #include <stdio.h> #include <dl.h> extern int errno; main() { shl_load("lib_not_found", BIND_DEFERRED, 0); printf("Error %d, %s ", errno, strerror(errno)); } $ cc a.c -lc -lpthread $ a.out Error 22, Invalid argument $ LD_PRELOAD=/usr/lib/libpthread.1 ./a.out Error 2, No such file or directory $ cat b.c #include <stdio.h> #include <dlfcn.h> void* handle; extern int errno; main() { handle = dlopen("lib_not_found", RTLD_LAZY); printf("Error %d, %s ", errno, strerror(errno)); if (handle == NULL) { printf("Error: %s ",dlerror()); } } $ cc b.c -lc -lpthread $ a.out Error 22, Invalid argument Error: $ ./a.out $ LD_PRELOAD=/usr/lib/libpthread.1 Error 0, Error 0 Error: Can't open shared library: lib_not_found Due to the problems mentioned above, should never be specified in the build command of an executable or shared library. By default, the compiler drivers automatically pass to the linker at the end of the link line of the executables. To see if a shared library was built with look at the shared library list in the output (see chatr(1)), or list the dependent libraries with (see ldd(1)): $ cc +z -c lib1.c $ ld -b -o lib1.sl lib1.o -lc $ ldd lib1.sl /usr/lib/libc.2 => /usr/lib/libc.2 /usr/lib/libdld.2 => /usr/lib/libdld.2 /usr/lib/libc.2 => /usr/lib/libc.2 $ cc +DA2.0W +z -c lib1.c $ ld -b -o lib1.sl lib1.o -lc $ ldd lib1.sl libc.2 => /lib/pa20_64/libc.2 libdl.1 => /usr/lib/pa20_64/libdl.1 To see the order in which dependent shared libraries will be loaded at run-time (order is only valid in 64-bit mode), use the command on the executable in 32-bit mode displays the order in which libraries are loaded in reverse order): $ cc +DA2.0W thread.c -lpthread $ ldd a.out libpthread.1 => /usr/lib/pa20_64/libpthread.1 libc.2 => /usr/lib/pa20_64/libc.2 libdl.1 => /usr/lib/pa20_64/libdl.1 $ cc +DA2.0W thread.c -lc -lpthread $ ldd a.out libc.2 => /usr/lib/pa20_64/libc.2 libpthread.1 => /usr/lib/pa20_64/libpthread.1 libdl.1 => /usr/lib/pa20_64/libdl.1 $ cc +DA2.0W thread.c -lpthread -lc $ ldd a.out libpthread.1 => /usr/lib/pa20_64/libpthread.1 libc.2 => /usr/lib/pa20_64/libc.2 libdl.1 => /usr/lib/pa20_64/libdl.1 Recommendations: o Remove from the build command of all shared libraries o Remove from the build command of all executables o Use the environment variable set to the full pathname for or which will cause the library to be loaded at program startup before other dependent libraries. functionality is available in and later Linker patches. See the dld.sl(5) man page. o If you link directly with the ld(1) command instead of with a compiler driver, add as the last component on the link line. Example 3 (64-bit) If a 64-bit shared library is built with but the executable is not, is loaded before (due to breadth-first searching), and the pthread calls are resolved to the pthread stubs in libc. At run-time, after the is loaded, the dependencies of are loaded in breadth-first order: is loaded as a dependent of before is loaded as a dependent of The dependency list of the first case is: a.out / / lib1 lib2 libc | | libc libpthread Therefore the load graph is constructed as: This is the desired behavior for non-threaded applications, but causes threaded applications (that use either or to fail. specifies specifies and no on $ cc -c +z +DA2.0W lib1.c lib2.c lib1.c: lib2.c: $ ld -b -o lib1.sl -lc lib1.o $ ld -b -o lib2.sl -lpthread lib2.o $ cc +DA2.0W thread.c -L. -l1 -l2 $ a.out Error $ ldd a.out lib1.sl => ./lib1.sl lib2.sl => ./lib2.sl libc.2 => /usr/lib/pa20_64/libc.2 libc.2 => /lib/pa20_64/libc.2 libpthread.1 => /lib/pa20_64/libpthread.1 libdl.1 => /usr/lib/pa20_64/libdl.1 specifies and no on $ ld -b -o lib1.sl lib1.o $ ld -b -o lib2.sl -lpthread lib2.o $ cc +DA2.0W thread.c -L. -l1 -l2 $ a.out Error $ ldd a.out lib1.sl => ./lib1.sl lib2.sl => ./lib2.sl libc.2 => /usr/lib/pa20_64/libc.2 libpthread.1 => /lib/pa20_64/libpthread.1 libdl.1 => /usr/lib/pa20_64/libdl.1 The same problem will occur if is listed as a dependent library of a shared library, and you would need to link the executable with Recommendation for Example 3 For threaded applications, run the executable with set to the library or link the executable with Use to load first $ ld -b -o lib1.sl lib1.o $ ld -b -o lib2.sl -lpthread lib2.o $ cc +DA2.0W thread.c -L. -l1 -l2 $ a.out Error $ ldd a.out lib1.sl => ./lib1.sl lib2.sl => ./lib2.sl libc.2 => /usr/lib/pa20_64/libc.2 libpthread.1 => /lib/pa20_64/libpthread.1 libdl.1 => /usr/lib/pa20_64/libdl.1 $ LD_PRELOAD="/lib/pa20_64/libpthread.1" a.out Success correctly lists for a threaded application. $ ld -b -o lib1.sl lib1.o $ ld -b -o lib2.sl -lpthread lib2.o $ cc +DA2.0W thread.c -L. -l1 -l2 -lpthread $ a.out Success $ ldd a.out lib1.sl => ./lib1.sl lib2.sl => ./lib2.sl libpthread.1 => /usr/lib/pa20_64/libpthread.1 libc.2 => /usr/lib/pa20_64/libc.2 libpthread.1 => /lib/pa20_64/libpthread.1 libdl.1 => /usr/lib/pa20_64/libdl.1 Example 4 (archived libc) If the link line of your shared library contains to explicitly link in remove Otherwise, shared libraries may be referencing while the may refer to older (archived) version. Thus the application will actually be using two different versions of and possibly mixing the code. This may cause compatibility problems. Basically, an application or library should never directly link against All programs need to be linked against (which the compiler does automatically), so a shared library will always have the interfaces it needs to execute properly without needing to specify on the link line. SEE ALSO
chatr(1), ld(1), ldd(1), pthread(3T), shl_load(3X), dld.sl(5). pthread_stubs(5)
All times are GMT -4. The time now is 05:03 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy