Control Transfer Termination Instructions for ROP Attack Prevention

The patent applies preliminary action by inserting CTT instructions at compile time into specific control transfer locations (function prologues, epilogues, exception handlers). These instructions are prepared in advance to enforce security policies, so when ROP attacks attempt to execute unintended control transfers, the pre-placed CTT instructions block them. This resolves the contradiction by establishing security constraints beforehand without requiring complex runtime analysis or validation mechanisms.

The patent applies local quality by selectively placing CTT instructions only at specific critical control transfer locations identified through compiler analysis, rather than uniformly at all control transfer points. The compiler identifies locations such as function prologues, epilogues, and exception handlers where control transfers occur, and inserts CTT instructions only at these locations. This selective approach maintains security against ROP attacks while minimizing the number of CTT instructions executed, thereby reducing performance overhead.

The patent applies the taking out principle by extracting and eliminating unnecessary control transfer tracking from the system. Instead of implementing comprehensive tracking of all control transfers, the approach removes tracking requirements for control transfers that are already validated by the presence of CTT instructions at critical locations. The compiler analysis identifies which control transfers need enforcement, and only those are tracked and validated by CTT instructions, while other control transfers proceed without overhead.