Abstract: A method for optimally issuing instructions that are related to a first instruction in a data processing system is disclosed. The processing system includes a primary and secondary cache. The method and system comprises speculatively indicating a hit of the first instruction in a secondary cache and releasing the dependent instructions. The method and system includes determining if the first instruction is within the secondary cache. The method and system further includes providing data related to the first instruction from the secondary cache to the primary cache when the instruction is within the secondary cache. A method and system in accordance with the present invention causes instructions that create dependencies (such as a load instruction) to signal an issue queue (which is responsible for issuing instructions with resolved conflicts) in advance, that the instruction will complete in a predetermined number of cycles.

Abstract: A method and apparatus for transporting store requests between functional units within a processor is disclosed. A data processing system includes a data dispatching unit, a data receiving unit, a segmented data pipeline coupled between the data dispatching unit and the data receiving unit, and a segmented feedback line coupled between the data dispatching unit and the data receiving unit. Having multiple latches interconnected between segments, the segmented data pipeline systolically transfers data from the data dispatching unit to the data receiving unit. The segmented feedback line has multiple control latches interconnected between segments. Each of the control latches sends a control signal to a respective one of the latches in the segmented instruction pipeline to forward data to a next segment within the segmented data pipeline.

Abstract: A method of storing values in a sliced cache by providing separate, but coordinated, reservation units for each cache slice. When a load-with-reserve (larx) operation is issued from the processor core as part of an atomic read-modify-write sequence, a message is broadcast to each of the cache slices to clear reservation flags in the slices; a reservation flag is also set in the target cache slice, and a memory address associated with the load-with-reserve operation is loaded into a reservation unit of the target cache slice. When a conditional store operation is issued from the core to complete the atomic read-modify-write sequence, a second message is broadcast to any non-target cache slice of the processing unit to clear reservation flags in the non-target cache slice(s).

Abstract: The present invention provides a system and method for efficient execution of load reserve (LARX) and store conditional (STCX) instructions in a superscalar processor. A system for efficiently providing a LARX instruction in a superscalar processor is disclosed. The system comprises a data cache (Dcache) for receiving the LARX instruction. The data cache further includes a decoder means for setting and resetting of a validation of the load reserve instruction, an internal cache for receiving address information and for providing data. The system also includes a register means for receiving the LARX instruction and a controller means for providing a physical address based upon the address information. The system provides for the validation being accomplished in one cycle for the LARX instruction when there is a hit on the internal data cache.