Since existing predictions were added many new opcodes and opcode combinations were added. Predictions need to be updated to reflect current status.

The patch contained only predictions for highly predictive (>=94%) pairs of commonly used opcodes or 100% predictive pairs of uncommonly used opcodes (but they can be more common in specific applications).

The second version of the patch contains only 100% predictions. It doesn't depend on any statistics, the compiler always generate these opcodes in pairs.