openSUSE Programming June 2007

programming@lists.opensuse.org

9 participants
4 discussions

[opensuse-programming] POSIX Thread programming question

by Romeo AHOHE

Hi, I have a question about POSIX threads programming regarding the program below. Basically, I have 3 threads that are doing nothing but printing a message and a 4th thread which continuously loop until all the other 3 threads terminate. To check that the other 3 threads terminate properly, the 4th thread make calls to pthread_getschedparam(). I tested the code on a machine running FreeBSD 5.3 and everything works fine. However, the behavior of the program is completely different on machines running Linux or Cygwin. The problem on those machines is that pthread_getschedparam() does not return the proper error code (ESRCH) even when the thread id that was passed to it is no longer running. I did notice however, that when I moved the code for the 4th process into main(), everything works fine, which seems to suggest that only parent process can access their child threads' scheduling state. So, my questions are: 1) In the POSIX thread implementation on Linux/Cygwin, can threads running within the same process access the scheduling state of their peers via pthread_getschedparam()? 2) Am I just doing something stupid by calling pthread_getschedparam() within my 4th thread ? Below is the code in question: #include <stdio.h> #include <pthread.h> #include <sched.h> #include <sys/errno.h> #define NUM_RECEIVERS 3 pthread_t rxThread[NUM_RECEIVERS]; pthread_t computeThread ; void *rxThreadFunc(void *arg) { int *i ; i = (int *)arg ; printf("Inside Thread %d\n", *i); return NULL ; } void *computeThreadFunc(void *arg) { int p[NUM_RECEIVERS] ; int k; int flag ; struct sched_param s[NUM_RECEIVERS] ; while(1) { flag = 1 ; for( k=0; k < NUM_RECEIVERS; k++ ) { if( pthread_getschedparam(rxThread[k], &(p[k]), &(s[k])) == ESRCH ) continue ; else { printf("Policy = %d\n", p[k]); flag = 0 ; break ; } } if( flag == 1 ) { printf("ALL THREADS HAVE STOPPED\n") ; break ; } else { printf("SOME THREADS ARE STILL RUNNING\n"); } } return NULL ; } int main() { int j, k; int p ; struct sched_param s ; for( j=0; j < NUM_RECEIVERS; j++ ) { if( pthread_create(&(rxThread[j]), NULL, rxThreadFunc, &j) != 0 ) { printf("Thread #%d was not created\n", j); for( k=0; k < j; k++ ) pthread_cancel(rxThread[j]); } pthread_join(rxThread[j], NULL) ; } if( pthread_create(&computeThread, NULL, computeThreadFunc, NULL) != 0 ) { printf("computeThread not created :-(\n"); for( k=0; k < NUM_RECEIVERS; k++ ) pthread_cancel(rxThread[j]); } pthread_join(computeThread, NULL); return 0 ; } ____________________________________________________________________________________ Get the free Yahoo! toolbar and rest assured with the added security of spyware protection. http://new.toolbar.yahoo.com/toolbar/features/norton/index.php --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-programming+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-programming+help(a)opensuse.org

14 years, 5 months

[opensuse-programming] debug / optimized / configure - what takes what

by opensuse＠sky-bizz.com

Hi there when i compile my kdevelop-projekt (simple C++ in console-mode) in "debug"-mode, i get a filesize of 300 KB, because of the debug-code insight. when i compile the same to "optimized"-mode, i get a smaller filesize of the executable of 40 KB. for making a dist i copied the whole project-folder to a temporary directory and run "make dist". then i was testing this distribution: i unpacked this zipped tarball to a directory, called there "./configure", "make" and "make install". everything worked fine. but: the compiled and installed executable now has a size of 170 KB, so this is neither the debug-mode nor the optimized-mode. what should i do, that the dist-copy would compiled like in "optimized"-mode from kdevelop? what compiler- and linker-options could be within the dist? thank you very much for any help! kind regards hagen --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-programming+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-programming+help(a)opensuse.org

14 years, 5 months

[opensuse-programming] 32-bit kernel module compilation

by Oliver Block

Hello list, I'd like to compile a 32-bit kernelmodule (ndiswrapper) on a 64-bit (AMD) System. My idea is to use the 32-bit runtime environment because I've got only a 32-bit Driver for a wlan card. The manufacturer is not planning to release a 64-bit driver for that card. Will my idea work and where do I find information about compiling 32-bit software on a 64-bit linux box using gcc? Best Regards, Oliver --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-programming-de+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-programming-de+help(a)opensuse.org --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-programming+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-programming+help(a)opensuse.org

14 years, 5 months

[opensuse-programming] Another C++ question

by Jerry Feldman

I have a static string defined in a class: class Foo { public: ... static const std::string default_name; ... }; In the C++ file: const std::string Foo::default_name("standard"); The C++ file is compiled into a shared library, say libfubar.so The executable is built: g++ main.cc -o fubar -L<path to libraries> -lfubar In the real case I have about a dozen libraries. To execute the code, I manually update LD_LIBRARY_PATH to point to the shared libraries including the RogueWave STL implementation, but I segfault in std::string when it is accessing Foo::default_name. I have tested with and without const. The backtrace shows that we are in the default constructor, Foo(). As a work around, I could possibly use a c language string (eg. char *), or even simply define the string as a manifest constant, but my preference is to use the STL string. -- Jerry Feldman <gaf(a)blu.org> Boston Linux and Unix user group http://www.blu.org PGP key id:C5061EA9 PGP Key fingerprint:053C 73EC 3AC1 5C44 3E14 9245 FB00 3ED5 C506 1EA9

14 years, 5 months

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

openSUSE Programming June 2007