#GTThreads — a preemptive user-level thread package with an API similar to Pthreads Zizheng Wu

Implemented APIs

void gtthread_init(long period);
int  gtthread_create(gtthread_t *thread,
                     void *(*start_routine)(void *),
                     void *arg);
int  gtthread_join(gtthread_t thread, void **status);
void gtthread_exit(void *retval);
int  gtthread_yield(void);
int  gtthread_equal(gtthread_t t1, gtthread_t t2);
int  gtthread_cancel(gtthread_t thread);
gtthread_t gtthread_self(void);

int  gtthread_mutex_init(gtthread_mutex_t *mutex);
int  gtthread_mutex_lock(gtthread_mutex_t *mutex);
int  gtthread_mutex_unlock(gtthread_mutex_t *mutex);

What Linux platform do you use?

The code is developed under Xubuntu.

Distributor ID: Ubuntu
Description:  Ubuntu 12.04.5 LTS
Release:  12.04
Codename: precise

How the preemptive scheduler is implemented.

SIGVTALRM is raised by a timer regarding the interval passed to gtthread_init. On cathing SIGVTALRM, the scheduler will perform swapcontext between the current thread and the next thread in queue.

How to compile your library and run your program.

Specifically, I have attached a solution to the Dining Philosophers problem using my implementation of GTThreads.

To run, make && ./dining_main .

A picture:

How you prevent deadlocks in your Dining Philosophers solution.

I applied the resource hierarchy solution. I numbered all the chopsticks. Every philosopher must pickup the lower-numbered fork first, and then the higher-numbered fork.

Any thoughts you have on the project, including things that work especially well or which don't work.

• good: I believe I gave an elegant implementation. Context swapping is performed by alrm_handler only. Every time when there is a need to swapcontext, I just raise(SIGVTALRM);. This will keep the logic of the program clear.
• bad: The program hasn't been tested thouroughly though it passed all the tests on Udacity. I suspect that boundary cases might break my library.