Improving cache locality for thread-level speculation

With the advent of chip-multiprocessors (CMPs), Thread-Level Speculation (TLS) remains a promising technique for exploiting this highly multithreaded hard-ware to improve the performance of an individual pro-gram. However, with such speculatively-parallel exe-cution the cache locality once enjoyed by the original uniprocessor execution is significantly disrupted: for TLS execution on a four-processor CMP, we find that the data-cache miss rates are nearly four-times those of the uniprocessor case, even though TLS execution utilizes four private data caches (i.e., four-fold greater cache capacity). We break down the TLS cache locality problem into instruction and data cache, execution stages, and par-allel access patterns, and propose methods to improve cache locality in each of these areas. We find that for parallel regions across 13 SPECint applications our simple and low-cost techniques reduce data-cache misses by 38%, improve performance by 12.8%, and significantly improve scalability—further enhancing the feasibility of TLS as a way to capitalize on future CMPs.