Abstract: A method and instruction for converting a number between a floating point format and an integer format are described. Numbers are stored in the integer format in a register of a first set of architectural registers in a packed format. At least one of the numbers in the integer format is converted to at least one number in the floating point format. The numbers in the floating point format are placed in a register of a second set of architectural registers in a packed format.

Abstract: A method and instruction for converting a number between a floating point format and an integer format are described. Numbers are stored in the integer format in a register of a first set of architectural registers in a scalar format. At least one of the numbers in the scalar format is converted to a number in the floating point format. The number in the floating point format is placed in a register of a second set of architectural registers in a packed format.

Abstract: A method and apparatus including conditional add and conditional add/subtract instructions are provided for use in the instruction set of a medical device instruction processor. More specifically, the conditional add and add/subtract instructions are provided to add two operands if a predetermined condition is satisfied within the instruction processor hardware. Additionally, the conditional add/subtract instruction may be used to subtract one operand from another operand if the predetermined condition is not satisfied. These instructions are adapted for use in implementing an efficient, interruptible, firmware-controlled multiplication or division mechanism. The inventive system allows multiplication or division operations to be interrupted at various intermediate points during the multiplication or division operation to thereby reducing interrupt latency.

Abstract: A method and system is disclosed which summarizes the results of a classical single-instruction multiple-data SIMD predicate comparison operation, signaling whether all comparisons resulted in a false result or true result, and placing that status into a separate status register, such as the Power PC Condition Register. The method and system utilizes first and second status bits to support the signaling whether all element comparisons resulted in true or false. The first status bit is set when all element comparisons resulted in false (i.e. a NOR of all predicate comparison results), and the second status bit is set when all element comparisons resulted in true (i.e. an AND of all predicate comparison results). This capability allows control flow using conditional branching on the event when all comparison results are false or when all comparison results are true.

Abstract: A method and apparatus for performing complex arithmetic is disclosed. In one embodiment, a method comprises decoding a single instruction, and in response to decoding the single instruction, moving a first operand occupying lower order bits of a first storage area to higher order bits of a result, moving a second operand occupying higher order bits of a second storage area to lower order bits of the result, and negating one of the first and second operands of the result.

Abstract: A plurality of units identified by respective addresses are disposed in a register file. Each of the units has a data register for storing data representative of a result of an arithmetic operation, a register for storing an overflow flag indicating the presence or absence of an occurrence of overflow in the arithmetic operation, and a register for storing a sign flag providing an indication of which one of a positive and a negative saturation value should replace the arithmetic operation result in the presence of an occurrence of overflow in the arithmetic operation. Each of the flags is updated in response to a write signal concerning its corresponding register. If at the moment when a data register is fed a read signal, a corresponding overflow flag is set, then either a positive or a negative saturation value, whichever corresponds to the sign flag, is generated on the input side of an arithmetic unit.

Abstract: A method and instruction for converting a number between a floating point format and an integer format are described. Numbers are stored in the integer format in a register of a first set of architectural registers in a packed format. At least one of the numbers in the integer format is converted to at least one number in the floating point format. The numbers in the floating point format are placed in a register of a second set of architectural registers in a packed format.

Abstract: A single chip microprocessor has at least two parallel pipelines that each have multiple processing stages, one of which is an instruction execution stage with a full functioned arithmetic logic unit (ALU). The ALU of one pipeline includes an adder that has the usual two input ports while the adder of the ALU of the other pipeline has at least one extra input port. Two successive arithmetically data dependent instructions are executed by the larger adder alone, while the smaller adder is used as part of a logic circuit that determines the carry bit for the instruction execution result obtained from the larger adder. The smaller adder is thus efficiently used, in an operation where it would otherwise be idle. The additional logic circuitry necessary to determine the carry bit is thus minimized.

Abstract: In addition to a register file having four general-purpose registers each for storing data, an arithmetic and logic unit for executing an addition instruction, a subtraction instruction, or the like, and a multiplier unit for executing a multiplication instruction, there are provided a controller and a substitute register for storing only data representing the result of operation performed by the multiplier unit in place of any of the four general-purpose registers in the register file. The controller controls the writing and reading of data in and from the register file and the writing and reading of data in and from the substitute register based on a multiplication tag indicative of the one of the four general-purpose registers in place of which the substitute register stores the data representing the result of multiplication and on a multiplication execute flag indicative of whether the data stored in the substitute register is effective or ineffective.

Abstract: An apparatus and method for bi-directional format conversion and transfer of data between integer and floating point registers is provided. A floating point register is configured to store floating point data, and integer data, in a variety of numerical formats. Data is moved in and out of the floating point register as integer data, and is converted into floating point format as needed. Separate processor instructions are provided for format conversion and data transfer to allow conversion and transfer operations to be separated.

Abstract: An information processing apparatus such as a microcomputer consisting of a CPU and a coprocessor is provided. The CPU and the coprocessor are connected through a data bus and an address bus. Switches are disposed in the data bus and the address bus which block communication between the CPU and the coprocessor upon execution of an instruction in the coprocessor, thereby allowing the CPU 1 to operate in parallel to the coprocessor.

Abstract: A method and instruction for converting a number from a floating point format to an integer format are described. Numbers are stored in the floating point format in a register of a first set of architectural registers in a packed format. At least one of the numbers in the floating point format is converted to at least one 16-bit number in the integer format. The 16-bit number in the integer format is placed in a register of a second set of architectural registers in the packed format.

Abstract: An optimized, superscalar microprocessor architecture for supporting graphics operations in addition to the standard microprocessor integer and floating point operations. A number of specialized graphics instructions and accompanying hardware for executing them are disclosed to optimize the execution of graphics instruction with minimal additional hardware for a general purpose CPU.

Abstract: A method and apparatus for performing a divide operation in a computer are described. The apparatus includes a first memory containing estimated reciprocal terms, and a second memory containing reciprocal error terms. An adder adds a selected estimated reciprocal term from the first memory and a selected reciprocal error term from the second memory to provide the reciprocal. The selected estimated reciprocal term and the selected reciprocal error term correspond to at least a portion of a divisor. The apparatus includes a multiplier for multiplying a dividend by the reciprocal to generate a quotient. The method includes the step of looking up an estimated reciprocal term in a first lookup table stored in a first computer memory wherein the estimated reciprocal term corresponds to at least a portion of a given divisor. A reciprocal error term is looked up in a second lookup table stored in a second computer memory, the error term corresponds to at least a portion of the divisor.

Abstract: A processor which executes positive conversion processing, which converts coded data into uncoded data, and saturation calculation processing, which rounds a value to an appropriate number of bits, at high speed. When a positive conversion saturation calculation instruction “MCSST D1” is decoded, the sum-product result register 6 outputs its held value to the path P1. The comparator 22 compares the magnitude of the held value of the sum-product result register 6 with the coded 32-bit integer “0x0000_00FF”. The polarity judging unit 23 judges whether the eighth bit of the value held by the sum-product result register 6 is “ON”. The multiplexer 24 outputs one of the maximum value “0x0000_00FF” generated by the constant generator 21, the zero value “0x0000_0000” generated by the zero generator 25, and the held value of the sum-product result register 6 to the data bus 18.

Abstract: A method and apparatus are disclosed for staggering execution of an instruction. According to one embodiment of the invention, a single macro instruction is received wherein the single macro instruction specifies at least two logical registers and wherein the two logical registers respectively store a first and second packed data operands having corresponding data elements. An operation specified by the single macro instruction is then performed independently on a first and second plurality of the corresponding data elements from said first and second packed data operands at different times using the same circuit to independently generate a first and second plurality of resulting data elements. The first and second plurality of resulting data elements are stored in a single logical register as a third packed data operand.

Abstract: An objective of this invention is a design that improves the memory usage ratio and execution speed of a sum-of-products operation instruction, improves the critical path of sum-of-products operations, and prevents overflows. A sum-of-products operation circuit executes sum-of-products operations a number of times that is specified by number-of-executions information comprised within a sum-of-products operation instruction, under the control of a control circuit. The number of times the sum-of-products operation is to be executed is set into a register, that number is decremented every time one cycle of the sum-of-products operation ends, and the sum-of-products operation instruction ends when the value in the register reaches zero. If an interrupt is received during the execution of a plurality of sum-of-products operations, execution of the sum-of-products operations resumes after the interrupt processing. First and second sum-of-products input data are read at the same time by a single memory access.

Abstract: The present invention generally relates to multiply-accumulate units for use in digital signal processors. Each multiply-accumulate unit includes a multiply unit which is coupled with two or more dedicated accumulators. Because of the coupling configuration, when an instruction specifies which accumulator should be used in executing an operation, the instruction need not specify which multiply unit should be utilized. A scheduler containing a digital signal processor's coupling configuration may then identify the multiply unit associated with the accumulator and may then forward the instruction to the identified multiply unit. Multiply-accumulate units can be configured to execute both scalar and vector operations. For executing vector operations, multiply units and their coupled accumulators are configured such that each may be easily grouped with other multiply units and accumulators.

Abstract: A method and apparatus are disclosed for staggering execution of an instruction. According to one embodiment of the invention, a single macro instruction is received wherein the single macro instruction specifies at least two logical registers and wherein the two logical registers respectively store a first and second packed data operands having corresponding data elements. An operation specified by the single macro instruction is then performed independently on a first and second plurality of the corresponding data elements from said first and second packed data operands at different times using the same circuit to independently generate a first and second plurality of resulting data elements. The first and second plurality of resulting data elements are stored in a single logical register as a third packed data operand.

Abstract: A system and method for determining an age function by performing a logical function on each entry residing within a queue, determining when a particular one of the entries residing in the queue was stored in the queue relative to the other entries, and determining an oldest or youngest entry residing in the queue relative to the logical functions performed on each of the instructions. In one embodiment of the present invention, the entries are instructions temporarily stored within a queue in the processor. The logical function performed may determine which of the instructions is valid. The queue may be cyclical.

Abstract: A multiplier capable of performing complex iterative calculations such as division and square root concurrently with simple independent multiplication operations is disclosed. The division and square root operations are performed using iterative multiplication operations such as the Newton Raphson iteration and series expansion. These iterative calculations may require a number of passes through the multiplier. Since the multiplier may be pipelined, it may experience a number of idle cycles during the iterative calculations. The multiplier is configured to utilize these idle cycles to perform independent simple multiplication operations. The multiplier may be configured to assert a control signal that is indicative of future idle cycles in the first stages of the multiplier pipeline. The control signal may be used by control logic to dispatch independent simple multiplication operations to the multiplier for execution during the idle clock cycles.

Abstract: An apparatus and method for calculating a square root of an operand in a microprocessor are provided. The microprocessor has a plurality of square root instructions, each of which specifies a square root calculation precision. The apparatus includes translation logic and execution logic. The translation logic decodes the square root macro instruction into a plurality of prescribed-precision machine instructions according to the square root calculation precision specified by the plurality of square root instructions. The execution logic, coupled to the translation logic, receives the plurality of prescribed-precision machine instructions and calculates the square root of the operand according to the specified square root calculation precision. At least one of the plurality of square root instructions specifies the square root calculation precision such that less significant bits are calculated in the square root than are provided in the operand.

Abstract: A data processing apparatus includes plural data registers, an arithmetic logic unit and a status register. The status register stores a plurality of different types of status bits. These status bits could be a negative status bit, a carry status bit, an overflow status bit and a zero status bit. These status bits are normally set dependent upon the condition of the result generated by the current arithmetic logic unit operation. A status bit protect instruction type permits selection of status bits protected from modification corresponding to the current arithmetic logic unit result. This status bit protect instruction preferably includes individual protect bit corresponding to each status bit. If a protect bit has a first digital state, then the corresponding status bit may be modified corresponding to the current arithmetic logic unit result.

Abstract: A multimedia extension unit (MEU) is provided for performing various multimedia-type operations. The MEU can be coupled either through a coprocessor bus or a local CPU bus to a conventional processor. The MEU employs vector registers, a vector ALU, and an operand routing unit (ORU) to perform a maximum number of the multimedia operations within as few instruction cycles as possible. Complex algorithms are readily performed by arranging operands upon the vector ALU in accordance with the desired algorithm flowgraph. The ORU aligns the operands within partitioned slots or sub-slots of the vector registers using vector instructions unique to the MEU. At the output of the ORU, operand pairs from vector source or destination registers can be easily routed and combined at the vector ALU.

Abstract: A single-chip multiprocessor (2, 102) is disclosed. The multiprocessor (2, 102) includes multiple central processing units, or CPUs, (10, 110) that share a floating-point unit (5, 105). The floating-point unit (5, 105) may receive floating-point instruction codes from either or both of the multiple CPUs (10, 110) in the multiprocessor (2, 102), and includes circuitry (52) for decoding the floating-point instructions for execution by its execution circuitry (65). A dispatch unit (56) in the floating-point unit (5, 105) performs arbitration between floating-point instructions if more than one of the CPUs (10, 110) is forwarding instructions to the floating-point unit (5, 105) at the same time. Dedicated register banks, preferably in the form of stacks (60), are provided in the floating-point unit (5, 105).

Abstract: Custom operations are useable in processor systems for performing functions including multimedia functions. These custom operations enhance a system, such as PC system, to provide real-time multimedia capabilities while maintaining advantages of a special-purpose, embedded solution, i.e., low cost and chip count, and advantages of a general-purpose processor reprogramability. These custom operations work in a computer system which supplies input data having at least two operand data, performs operations on the operand data, and supplies result data to a destination register.

Abstract: An improved digital signal processor, in which arithmetic multiply-add instructions are performed faster with substantial accuracy. The digital signal processor performs multiply-add instructions with look-ahead rounding, so that rounding after repeated arithmetic operations proceeds much more rapidly. The digital signal processor is also augmented with additional instruction formats which are particularly useful for digital signal processing. A first additional instruction format allows the digital signal processor to incorporate a small constant immediately into an instruction, such as to add a small constant value to a register value, or to multiply a register by a small constant value; this allows the digital signal processor to conduct the arithmetic operation with only one memory lookup instead of two.

Abstract: A microprocessor capable of unpacking packed data in response to an unpack instruction. The microprocessor having a a storage area to store a first packed data and a second packed data respectively including a first plurality of data elements and a second plurality of data elements, wherein each data element in the first plurality of data elements corresponds to a different data element in the second plurality of data elements, in a respective position. The microprocessor also includes a circuit that simultaneously copies less than all data elements from the first plurality of data elements and corresponding data elements from the second plurality of data elements into a storage area as a third plurality of separate data elements in a third packed data in response to the unpack instruction.

Abstract: A numerical processing method on a computer system in which an instruction having at least one operand and a type control is retrieved, and the operand is converted to a precision specified by the type control. The instruction is executed in the precision specified by the type control to obtain a result, and when the destination precision differs from the precision specified by the type control, the result is converted to the destination precision using a second instruction.

Abstract: A real time digital systolic processor with a core of reconfigurable interconnected macrocells which can be programmed according to function for processing high bandwidth digital data. Each macrocell contains arithmetic logic units for performing predetermined functions based on format of the input data stream from an outside source or from other macrocells. The interconnects between each macrocell are arranged so that the function of the device is predetermined according to user specific applications.

Abstract: A method for controlling a microprocessor to transform data from a signed format to an unsigned format so that the data can be processed by unsigned instructions. In particular, a subtraction between two signed numbers can be transformed into a subtraction between two unsigned numbers.

Abstract: A modular computational structure includes a pipeline having first and second adder stages. Each adder stage includes a pair of adders which operate in parallel, and outputs ports of the first adder stage are coupled to input ports of the second adder stage. Rounding logic and an accumulator are included in the second stage. By varying the inputs to the first and second stages a variety of complex arithmetic functions suitable for video encoding can be implemented. Examples of the operations include completion of multiply and multiply-and-accumulate operations, averages of two values, averages of four values, and merged difference and absolute value calculation.

Abstract: Along with an arithmetic processing unit and an arithmetic execution unit, another arithmetic processing unit is coupled in parallel to an instruction issue unit. Disposed within one of the arithmetic processing units are an address generation unit, an instruction buffer, an instruction decoder, an arithmetic execution unit, a data memory, and a flag register. The instruction decoder decodes an instruction read from the instruction buffer. If the decoded instruction is an iteration start instruction, the instruction decoder extracts a number of times an iterative process is to be executed that is included in the instruction for forwarding to the address generation unit. The address generation unit exerts control as to the execution and termination of iterative processes.

Abstract: An apparatus and method for performing integer division in a microprocessor are provided. The apparatus includes translation logic, floating point execution logic, and integer execution logic. The translation logic decodes an integer divide instruction into an integer divide micro instruction sequence and an overflow detection micro instruction sequence. The integer divide micro instruction sequence is routed to and executed by the floating point execution logic. The overflow detection micro instruction sequence is routed to and executed by the integer execution logic. The integer execution logic and the floating point execution logic execute the overflow detection micro instruction sequence and the integer divide micro instruction sequence concurrently.

Abstract: A method for scheduling a flag generating instruction and a subsequent instruction. The subsequent instruction has a data dependency on the flag generating instruction. The flag generating instruction is translated into first and second instructions. The subsequent instruction is translated into at least a third instruction. The first instruction, when executed, generates a result and intermediate flag generation data. The second instruction, when executed, generates a plurality of flags. The first instruction is scheduled to execute before the second and third instructions. The second instruction is scheduled to execute before the third instruction if the third instruction has a data dependency on the second instruction, otherwise the third instruction may be scheduled to execute before the second instruction.

Abstract: The present invention is a method and apparatus for reporting exception in a single instruction multiple data (SIMD) processor in computing an arithmetic function for a plurality of argument data. The SIMD processor is configured for processing N elements simultaneously. A sequence of instructions is re-arranged to allocate the plurality of argument data in the N elements. The N elements are processed simultaneously. The exceptions for N elements are detected simultaneously. The detected exceptions are then combined to generate a global exception.

Abstract: A system for processing a floating point instruction includes a stack, virtual registers, a stack pointer pointing to one of the virtual registers as top of stack, physical registers, and a reference table mapping the virtual registers to the physical registers, entries of the reference table pointing to physical register locations. An instruction unit generates a plurality of instructions, and a decode unit having a plurality of decoders receives the plurality of instructions from the instruction unit, respectively. The decode unit decodes the plurality of instructions and determines whether any one of the instructions contains a floating point instruction including a floating point exchange instruction. A logic unit is coupled to the reference table and includes a plurality of logic devices coupled to the plurality of decoders in the decode unit, respectively.

Abstract: A multiplier capable of performing both signed and unsigned scalar and vector multiplication is disclosed. The multiplier is configured for use in a microprocessor and comprises a partial product generator, a selection logic unit, and an adder. The multiplier is configured to receive signed or unsigned multiplier and multiplicand operands in scalar or packed vector form. The multiplier is also configured to receive a first control signal indicative of whether signed or unsigned multiplication is to be performed and a second control signal indicative of whether vector multiplication is to be performed. The multiplier is configured to calculate an effective sign for the multiplier and multiplicand operands based upon each operand's most significant bit and the control signal. The effective signs may then be used by the partial product generation unit and the selection logic to create and select a number of partial products according to Booth's algorithm.

Abstract: A method and apparatus to divide a signed integer by a constant power of two using conditionally-executed instructions to choose between a first result in the event that the dividend is a negative signed integer and a second result in the event that the dividend is a positive signed integer, wherein values associated with the first result and the second result are generated simultaneously.

Abstract: An arithmetic logic unit capable of executing an instruction belonging to a user-defined instruction area at the same clock frequency as a hard-wired logic includes a memory storing data at an arbitrary address and outputting the data stored in the address when an instruction code and an operand data are applied as an address. When an instruction decoder decoding part of the instruction code for setting the memory to read mode or write mode is provided, contents of the memory can be re-written, and therefore the content of the memory can be readily changed even after delivery. The arithmetic logic unit may include, in place of the memory, a programmable logic device adapted to receive an instruction code and the operand data and capable of organizing a desired logic.

Abstract: A microprocessor circuit is disclosed for instructions on an arithmetic/shift function performing standard operations (e.g., ALU instructions or Shift instructions) on instructions in a first cycle of operation, and a correction circuit responsive to the arithmetic/shift function for modifying the standard results provided by the arithmetic/shift function to results required by a SIMD instruction being executed. The arithmetic/shift function is an instruction provided by either an Arithmetic Logic Unit (ALU) or by a shift instruction. The correction circuit passes data, unchanged for logical instructions but provides condition codes according to the SIMD instruction.

Abstract: An intermediate result signal arising from a manipulation of data signals is checked and reduced without using conditional branches, thereby improving instruction processing. Data signals are represented as signed 8-bit binary values in a two's compliment format. This requires that the intermediate result signal be stored in a register that is greater than 8-bits wide to allow for the proper checking of an overflow condition. A processor operating under program control with the program has the following operations. The program determines whether the intermediate result signal is in a positive overflow state or a negative overflow state. A first mask signal is set to have 8 lower bits in an OFF position when the intermediate result signal is inside the range of a signed 8 bit integer. Otherwise, the first mask signal is set to have 8 lower bits in an ON position.

Abstract: An optimized, superscalar microprocessor architecture for supporting graphics operations in addition to the standard microprocessor integer and floating point operations. A number of specialized graphics instructions and accompanying hardware for executing them are disclosed to optimize the execution of graphics instruction with minimal additional hardware for a general purpose CPU.

Abstract: A routable operand and selectable operation processor multimedia extension unit is employed to motion compensate MPEG video using improved vector processing. A vector processing unit executes an add and divide instruction that adds two vector registers and divides the result in a single instruction. This is implemented through loading a first vector register with a first plurality of elements from a source block. A second vector register is then loaded with a second plurality of elements that are adjacent to the first plurality of elements. The add and divide instruction is then executed on the first and second vector registers, yielding an interpolated source element that is stored in a resultant vector register.

Abstract: A processor to coprocessor interface supporting multiple coprocessors utilizes compiler generatable software type function call and return, instruction execute, and variable load and store interface instructions. Data is moved between the processor and coprocessor on a bidirectional shared bus either implicitly through register snooping and broadcast, or explicitly through function call and return and variable load and store interface instructions. The load and store interface instructions allow selective memory address preincrementation. The bi-directional bus is potentially driven both ways on each clock cycle. The interface separates interface instruction decode and execution. Pipelined operation is provided by indicating decoded instruction discard by negating a decode signal before an execute signal is asserted.

Abstract: A processor which executes positive conversion processing, which converts coded data into uncoded data, and saturation calculation processing, which rounds a value to an appropriate number of bits, at high speed. When a positive conversion saturation calculation instruction "MCSST D1" is decoded, the sum-product result register 6 outputs its held value to the path P1. The comparator 22 compares the magnitude of the held value of the sum-product result register 6 with the coded 32-bit integer "0x0000.sub.-- 00FF". The polarity judging unit 23 judges whether the eighth bit of the value held by the sum-product result register 6 is "ON". The multiplexer 24 outputs one of the maximum value "0x0000.sub.-- 00FF" generated by the constant generator 21, the zero value "0x0000.sub.-- 0000" generated by the zero generator 25, and the held value of the sum-product result register 6 to the data bus 18.

Abstract: A three input arithmetic logic unit (230) generates a combination of the three inputs that is selected by a function signal. The second input signal comes from a controllable shifter (235). The shift amount is a default shift amount stored in a special data register, a predetermined set of bits of data recalled from a data register or zero. A one's constant source (236) is connected to the shifter (235) to supply a multibit digital signal of "1". This permits generating a second input signal of the form 2.sup.N, with N being the shift amount. The output of the shift (235) may be stored independently of the arithmetic logic unit (230) result. The third input signal comes from a multiplexer (233) that selects between an instruction specified immediate field, data recalled from a data register or a mask input from a mask generator (239). One preferred form of the mask has a number of right justified 1's corresponding to a mask input signal.

Abstract: A three input arithmetic logic unit (230) that generates a combination of the three inputs that is selected by a function signal. The second input signal comes from a controllable barrel rotator (235). The rotate amount is a default rotate amount stored in a special data register, a predetermined set of bits of data recalled from a data register or zero. A one's constant source (236) is connected to the barrel rotator (235) to supply a multibit digital signal of "1". This permits generating a second input signal of the form 2.sup.N, with N being the rotate amount. The output of the barrel rotator (235) may be stored independently of the arithmetic logic unit (230) result. The third input signal comes from a multiplexer (233) that selects between an instruction specified immediate field, data recalled from a data register or a mask input from a mask generator (239). One preferred form of the mask has a number of right justified 1's corresponding to a mask input signal.

Abstract: A processor having a modified Harvard architecture having first and second memories, an address register file divided into first and second sets of registers, first and second stream registers, and a general purpose register file for performing data streaming. The first and second sets of registers respectively address the first and second memories which, in turn, load data into the first and second stream registers. An arithmetic logic unit (ALU) accepts the stream registers and general purpose registers as inputs. Stream instructions are encoded such that a single instruction specifies an ALU operation performed on selected ALU inputs and where to store the results of the ALU operation, loads new values into the stream registers, and updates the address registers. A stream instruction has three operand fields respectively specifying two operands for the next ALU operation and a location to store the result of the current ALU operation.

Abstract: A CPU (central processing unit) of a computer that comprises an issue unit and an execution unit. The issue unit selectively issues arithmetic instructions of a predefined arithmetic instruction type as performance critical arithmetic instructions and non-performance critical arithmetic instructions. The execution unit comprises a performance critical pipeline to execute the performance critical arithmetic instructions. The execution unit also comprises a non-performance critical pipeline to execute the non-performance critical arithmetic instructions.