Abstract: A processor employing a post-cache (LS2) buffer. Loads are stored into the LS2 buffer after probing the data cache. The load/store unit snoops the loads in the LS2 buffer against snoop requests received from an external bus. If a snoop invalidate request hits a load within the LS2 buffer and that load hit in the data cache during its initial probe, the load/store unit scans the LS2 buffer for older loads which are misses. If older load misses are detected, a synchronization indication is set for the load misses. Subsequently, one of the load misses completes and the load/store unit transmits a synchronization signal with the status for the load miss. The processor synchronizes to the instruction corresponding to the load miss, thereby discarding load hit which was subsequently snoop hit. The discarding instructions are refetched and reexecuted, thereby causing the load hit to reexecute subsequent to an earlier load miss.

Abstract: A mechanism which manages variable length instructions in cache is comprised of three cooperating elements designed to optimize self modifying code and anticipate next instructions for branch operand management. A content addressable memory (CAM) stores addresses of lines which have been accessed for instruction fetching. In a system having modifiable instruction stream (i.e., store to instruction stream), when the CAM matches, the system must retire certain instructions, flush instructions and then fetch the modified instruction stream. Boundary identification logic examines a field in each cache byte to determine the nature of the byte. This field is initially cleared at the time the cache line is loaded and filled with the line is fetched. An anticipation buffer designed to minimize the circuitry necessary for fetches across cache lines is loaded with sequentially anticipated prefetched instructions from the cache. These anticipated instructions can then be concatenated by a fetch aligner.