Does Linux use some of the solutions of priority inversion?

Question

As known, priority inversion problem - when a thread with higher priority waits for a thread with lower-priority: https://en.wikipedia.org/wiki/Priority_inversion

It happen when we have 3 threads: L (low), M (medium), H (high-priority). L and H use the same mutex, but L acquire it early than H, and H blocked and goes to sleep. And then M occupies CPU-core because has higher priority than L, and L goes to sleep, but mutex still acquired. L & H are sleeping, but M is working.

There are some solutions of priority inversion:

• Disabling all interrupts to protect critical sections
• A priority ceiling
• Priority inheritance
• Random boosting - Ready tasks holding locks are randomly boosted in priority until they exit the critical section. This solution is used in Microsoft Windows.
• Avoid blocking

Does Linux use some of the solutions of priority inversion, and which of its?

Answer 1

Linux uses Priority inheritance.

Priority inversion can be solved by using function int pthread_mutexattr_setprotocol(pthread_mutexattr_t *attr, int protocol);

Where protocol is:

• PTHREAD_PRIO_NONE - nothing going on when acquires mutex ownership
• PTHREAD_PRIO_INHERIT - Priority inheritance
• PTHREAD_PRIO_PROTECT - uses a fixed level of the priority in accordance with prioceiling value which got by function pthread_mutexattr_getprioceiling()

When a thread owns one or more mutexes initialized with the PTHREAD_PRIO_PROTECT protocol, it shall execute at the higher of its priority or the highest of the priority ceilings of all the mutexes owned by this thread and initialized with this attribute, regardless of whether other threads are blocked on any of these mutexes or not.