Gedare Bloom started a new discussion on cpukit/posix/src/aio_misc.c: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/275#note_114375 > + pthread_mutex_lock( &suspendcbp->mutex ); > + > + if ( --suspendcbp->requests_left == 0 ) { > + finished = 1; > + } > + > + if ( suspendcbp->notified == AIO_NONSIGNALED ) { > + send_event = 1; > + suspendcbp->notified = AIO_SIGNALED; > + } > + > + pthread_mutex_unlock( &suspendcbp->mutex ); > + > + if ( send_event ) { > + rtems_event_system_send( > + suspendcbp->task_id, here you read from the `suspendcbp` outside of the lock. Is there a race condition here that allows another thread to reach `free(suspendcbp)` before this read occurs? -- View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/275#note_114375 You're receiving this email because of your account on gitlab.rtems.org.
_______________________________________________ bugs mailing list [email protected] http://lists.rtems.org/mailman/listinfo/bugs
