Reducing stalling in a simultaneous multithreading processor by inserting thread switches for instructions likely to stall

Abstract

A system and method suppresses occurrence of stalling caused by data dependency other than register dependency in an out-of-order processor. A stall reducing method includes a handler for detecting a stall occurring during execution of execution code using a performance monitoring unit, and for identifying, based on dependencies, a second instruction on which a first instruction is data dependent, the stall based on this dependency. A profiler registers the second instruction as profile information. An optimization module inserts a thread yield instruction in an appropriate position inside execution code or an original code file based on the profile information, and outputs optimized execution code.