1. Hyperthreading
... a fixed number of entries per logical processor simplifies design, versus allowing the entries to be dynamically allocated. Finally, resources like caches (trace and L1 data), branch global history array, microcode ROM, the scheduler, register renaming control logic, and execution units are fully shared, though not partitioned. The physical register files are shared and not hard partitioned (specific registers are not assigned ... the logical processors is stalled, full trace cache bandwidth is given to the running thread per cycle. In the less likely situation of both threads requiring...
- Word Count: 2124
- Approx Pages: 8
- Grade Level: Undergraduate