Why are synchronization mechanisms necessary in single-processor systems when working with multithreaded applications?

In single-processor systems, although only one thread is executed at any given moment, multithreaded applications can switch between threads (context switching). If multiple threads access shared resources (such as variables or data structures) simultaneously, race conditions can occur without synchronization mechanisms, leading to inconsistent data.

Synchronization mechanisms (mutexes, semaphores, locks) ensure that only one thread can modify or read a shared resource at a time, preventing errors and ensuring correct program operation.

Example: if two threads simultaneously increment a shared counter without locking, the final value may be less than expected due to conflicts during reading and writing.