https://issues.apache.org/bugzilla/show_bug.cgi?id=49504
--- Comment #11 from Yann Ylavic <[email protected]> --- (In reply to Yann Ylavic from comment #10) > Yet another possibility would be to not call pthread_mutex_destroy() in > proc_mutex_proc_pthread_cleanup(), only munmap()ing the region proc_mutex_proc_pthread_cleanup() could also destroy a copy of the pthread_mutex_t, eg. Index: locks/unix/proc_mutex.c =================================================================== --- locks/unix/proc_mutex.c (revision 1568989) +++ locks/unix/proc_mutex.c (working copy) @@ -347,7 +347,8 @@ static apr_status_t proc_mutex_proc_pthread_cleanu } /* curr_locked is set to -1 until the mutex has been created */ if (mutex->curr_locked != -1) { - if ((rv = pthread_mutex_destroy(mutex->pthread_interproc))) { + pthread_mutex_t copy = *mutex->pthread_interproc; + if ((rv = pthread_mutex_destroy(©))) { #ifdef HAVE_ZOS_PTHREADS rv = errno; #endif -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
