Critique of paper “ Multiplex : Unifying Conventional and Speculative Thread - Level

In this paper, the authors mention that there are two forms of TLP for Chip multiprocessors (CMPs): explicit threading and implicit threading. However, explicit threading is hard to program manually and, if automated, is limited in performance due to serialization of unanalyzable program segments. Implicit threading, on the other hand, requires buffering of program state to handle misspeculations, and is limited in performance due to buffer overflow in large threads and dependences in small threads. So this paper propose the Multiplex architecture to unify implicit and explicit threading in CMPs to combine the complementary strength of implicit and explicit threading to alleviate the individual weakness of these two schemes. The following are my comments on this paper: Strengths: