P2L2 - part 3

What is a common mistake when using mutex/condition variables?

• Forgetting to lock or unlock a mutex

• Using different mutexes to access a single resource

• Signaling that reads can occur instead of signaling that writes can occur

• Using signal instead of broadcast

What is a spurious wake-up?

• When a thread wakes up unnecessarily without being able to proceed

• When a thread wakes up and proceeds as expected

• When a thread wakes up and immediately goes back to sleep

• When a thread wakes up and causes a deadlock

What is a deadlock?

• When two or more threads are waiting on each other to complete

• When a thread wakes up unnecessarily without being able to proceed

• When a thread wakes up and proceeds as expected

• When a thread wakes up and immediately goes back to sleep

What is the recommended solution to avoid deadlocks?

• Unlock A before locking B

• Get all locks up front and release them at the end

• Maintain a lock order

• Do nothing and hope the system never deadlocks

What is the difference between kernel level threads and user-level threads?

• Kernel level threads are managed by the operating system, while user-level threads are managed by the user-level library

• Kernel level threads are visible to the kernel, while user-level threads are not

• Kernel level threads are more efficient than user-level threads

• Kernel level threads can only be used for system-level services

What is the one-to-one model of multithreading?

• Each user level thread has a kernel level thread associated with it

• All user level threads are mapped onto a single kernel level thread

• Some user level threads have a one-to-many relationship with a kernel thread

• Some user level threads have a many-to-one relationship with a kernel thread

What is the many-to-one model of multithreading?

• Each user level thread has a kernel level thread associated with it

• All user level threads are mapped onto a single kernel level thread

• Some user level threads have a one-to-many relationship with a kernel thread

• Some user level threads have a many-to-one relationship with a kernel thread

What is the many-to-many model of multithreading?

• Each user level thread has a kernel level thread associated with it

• All user level threads are mapped onto a single kernel level thread

• Some user level threads have a one-to-many relationship with a kernel thread

• Some user level threads have a many-to-one relationship with a kernel thread

What is the boss/workers pattern?

• A pattern where one boss thread assigns work to multiple worker threads

• A pattern where multiple boss threads assign work to one worker thread

• A pattern where one boss thread assigns work to one worker thread

• A pattern where multiple boss threads assign work to multiple worker threads

What is a benefit of the boss/workers pattern?

• Specialization and locality

• Increased throughput

• Simplified thread pool management

• Improved load balancing