In a data processing system in which the program memory unit and the internal data memory unit are separately addressed, and which special instructions for transferring data between these two units are available, apparatus and method are described for increasing the memory available for the internal data memory unit. The increased storage space takes the form of an auxiliary data memory unit that is activated when the special instruction for the transfer of data signal groups between the program memory unit and the data memory unit are evoked. Apparatus is provided for identifying the presence of the special instructions, and this identification generates the signals for inactivating the program memory unit and for activating the auxiliary data unit for the purposes of transfer of data signal groups between the internal data unit and the auxiliary data memory unit.
An extended Harvard architecture memory system which features an address store for containing an ordered sequence of program memory addresses, and a value store for containing a series of related data value sets. Each of the addresses contained in the address store is associated with a distinct set of instructions, such as a subroutine, that is contained in the program memory. The address store may also contain the address of one or more instruction arguments that are, in turn, contained in the value store or in a separate data memory. Both the address store and the value store are preferably connected to the same data communication path which is used by the data memory of the computer. The value store also includes a logic interface for enabling a plurality of different address increments to be programmably selected.
A massively multiplexed central processing unit ("CPU") which has a plurality of independent computational circuits, a separate internal result bus for transmitting the resultant output from each of these computational circuits, and a plurality of general purpose registers coupled to each of the computational circuits. Each of the general purpose registers have multiplexed input ports which are connected to each of the result buses. Each of the general purpose registers also have an output port which is connected to a multiplexed input port of at least one of the computational circuits. Each of the computational circuits are dedicated to at least one unique mathematical function, and at least one of the computational circuits include at least one logical function. At least one of the computational circuits includes a plurality of concurrently operable mathematical and logical processing circuits, and an output multiplexer for selecting one of the resultant outputs for transmission on its result bus. The CPU also features a very long instruction word which uses a series of assigned bit locations to represent the selections codes for each of the CPU components. These selection codes are directly transmitted to each of the CPU components by a program control circuit. A separate data control circuit is further provided in achieve a Harvard architecture design for the CPU.
