Abstract: A configuration of vector units, digital circuitry and associated instructions is disclosed for the parallel processing of multiple Viterbi decoder butterflies on a programmable digital signal processor (DSP) that is based on single-instruction-multiple-data (SIMD) principles and provides indirect access to vector elements. The disclosed configuration uses a processor with two vector units and associated registers, where the vector units are connected back to back for processing Viterbi decoder state metrics. Viterbi add instructions increment vectors of state metrics from a first register, performing a desired permutation of state metrics while reading them indirectly through vector pointers, and writing intermediate result vectors to a second register.

Abstract: A configuration of vector units, digital circuitry and associated instructions is disclosed for the parallel processing of multiple Viterbi decoder butterflies on a programmable digital signal processor (DSP) that is based on single-instruction-multiple-data (SIMD) principles and provides indirect access to vector elements. The disclosed configuration uses a processor with two vector units and associated registers, where the vector units are connected back to back for processing Viterbi decoder state metrics. Viterbi add instructions increment vectors of state metrics from a first register, performing a desired permutation of state metrics while reading them indirectly through vector pointers, and writing intermediate result vectors to a second register.

Abstract: Object-code compatibility is provided among VLIW processors with different organizations. The object-code can also be executed by sequential processors, thus providing compatibility with scalar and superscalar processors, A mechanism is provided which allows representing VLIW programs in an implementation-independent manner. This mechanism relies on instruction cache (I-cache) reload/access logic which incorporates implementation-dependent features into a VLIW program. In this way, programs are represented in main memory in an implementation-independent manner, the implementation-specific aspects are introduced as part of the instruction cache reload/fetch processes, and the simplicity in instruction dispatch logic that is characteristic of VLIW processors is preserved. The foregoing allows for object-code compatibility among VLIW processors with different organizations. Also provided is a mechanism and an apparatus for the interpretation of tree-instructions by a computer system based on a VLIW processor.

Abstract: The present invention is an apparatus that maps the memory address space of the computer system into regions, and detects the incorrect execution of a load operation performed earlier than a sequentially preceding (in program order) store operation. The apparatus detects out-of-order load operations, uses a region-based mapping table to keep track of the memory regions accessed by the out-of-order load operations, detects the execution of store operations into regions accessed by out-of-order load operations, and generates a program exception when interference among reordered operations is detected. The invention is applicable to static and dynamic reordering of memory operations.

Abstract: In a computer system having a hierarchical memory, the problem of tolerating cache miss latency is solved by dynamically switching appropriately between two different code sequences, one optimized at compile-time, assuming a cache-hit, and the other optimized at compile-time, assuming a cache-miss.

Abstract: A computer processing system stores sequences of instructions in a memory for execution by a processor unit. An out-of-order load instruction may be created, either statically or dynamically, by moving a load instruction from its original position in a sequence of instructions to an earlier position in said sequence of instructions. Such out-of-order load instruction identifies a location in memory from which to read a datum and a first destination register in which to place the datum. The present invention is a method and corresponding apparatus that utilizes data comparison to detect coherence among memory and the datum read by an out-of-order load operation. More specifically, the method consists of an interference test which controls the processor unit to read a datum from the same location in memory identified by the out-of-order load instruction and compare the newly read datum with the datum saved in the first destination register.

Abstract: Object code compatibility is provided among VLIW processors with different organizations. The object code can be executed by sequential processors, thus providing backward compatibility with scalar and superscalar processors. A mechanism is provided which allows representing VLIW programs in an implementation independent manner. This mechanism relies on instruction cache (I-cache) reload/access processes which incorporate implementation-dependent features into a VLIW program. In this way, programs are represented in main memory in an implementation independent manner (i.e., without reflecting the organization of the processor where they are executed), the implementation-specific aspects are introduced as part of the instruction cache reload/fetch processes, and the simplicity in instruction dispatch logic that is characteristic of VLIW processors is preserved. This allows for object code compatibility among VLIW processors with different organizations.