Abstract: A single shared processing path is used as contexts are switched during processing. Each unique context is processed using a corresponding unique pipeline. If a pipeline that is executing under one context stalls, processing is switched in the shared processing path to another pipeline that is executing under second context. New pipelines are enabled for execution by borrowing a clock cycle from the currently executing pipeline. In some cases contexts are assigned various relative priority levels. In one case a context switching microprocessor is used in a communication engine portion of a system-on-a-chip communication system.

Abstract: A controller for executing instructions has one the order of five addressing modes and can allow executing of processes concurrently in multiple modes. A specific embodiment can effectively run legacy code written for the Z80 micoprocessor without requiring recompiling of code. An optional embodiment includes autonomous Multiply/Accumulator Engine (MAC) optimized to perform sum-of-products (SOP) operations with little controller overhead, making the invention capable of more effectively handling a number of processing tasks, particularly tasks related to digital signal processing (DSP).

Abstract: An improved DMA controller is provided. The improved DMA controller uses a peripheral control bus which has scan codes to indicate the DMA channel, conventional data request/data acknowledge lines, and additional lines indicating a "terminate," "type fetch," "end of buffer" and "store status." List entries are associated with the buffers in the memory. Each list entry has a type/status field which can be coded with information indicating "ready buffer," whether to notify "end of buffer," "buffer in progress," "completed buffer without status," "completed buffer with status," "ready buffer with command," and "ready buffer without command." The type of status byte can be checked before processing the buffers.

Abstract: Delay circuitry is used in a circuit to delay the transmission of groups of data until another circuit expects these groups of data. In one embodiment, emulating circuitry is used to emulate the timing of transmitter and receiver UART FIFOs. This emulating circuitry uses delays equal to the amount of time the UART FIFOs take to serially shift out data in the transmitter UART FIFO, and to serially shift in data in the receiver UART FIFO. This allows the modem chip to use a parallel-to-parallel FIFO buffer for the transmitter FIFO buffer and the receiver FIFO buffer.

Abstract: A mask value generator is used to produce a mask value for deencrypting encrypted instructions from a memory. This encryption mask generator can produce a encryption mask value from a seed value and a program counter value. The seed value can stored in a memory outside the core microprocessor. The encryption mask value can be used to deencrypt instructions in an "exclusive or" logic section.