Branch prediction and store-to-load-forwarding are clearly not the only “optimistic” techniques in processors. We want to investigate all “optimistic” techniques across various processors (Intel, AMD, ARM, IBM, Apple, etc.) to find the next Spectre as early as possible. Any mechanism performing an operation not explicitly requested by the instruction stream must be analyzed for possible leakage. We are already working in this direction and would be happy to intensify our efforts here.
Taiga RIG Inventory project link here