Texas Instruments Corporation

Octavo Systems LLC

Laurence Ray Simar, Jr

Jerald G Leach

Natarajan Kurian Seshan

Reid E Tatge

Alan Davis

Gene A Frantz

Peter Robert Linder

Erik James Welsh

Joseph A Coomes

Steve P Marshall

Donald Edward Steiss

Richard H Scales

Nat Seshan

Richard A Brown

Michael D Beck

The present disclosure describes analog memories for use in a computer, such as a computer using a combination of analog and digital components/elements used in a cohesive manner.

Analog memory cells with valid flag

The present disclosure describes a mixed signal arithmetic logic unit configured to use a combination of analog processing elements and digital processing elements in a cohesive manner. Depending on the signals and the data received for processing, the analog processing elements and digital processing elements may be used separately, independently or in combination to optimize computational results and the performance of the mixed signal arithmetic logic unit.

Analog arithmetic unit

The present disclosure describes a mixed signal computer unit using a combination of analog and digital components/elements in a cohesive manner. Depending on the signals and data that need to be processed, the analog processing elements and digital processing elements may be used separately, independently or in combination to optimize the computational results and the performance of the computation. Operations may be controlled by one or more digital cores.

Mixed signal computer

A method of integrated circuit programmed data processor design includes selecting a benchmark application, selecting an initial set of architecture parameters, reconfiguring a compiler for the selected architecture parameters, compiling the benchmark, reconfiguring a data processor simulator to the selected architecture parameters, running the complied benchmark on the reconfigured simulator, automatically synthesizing an integrated circuit physical layout and evaluating performance of the selected architecture parameters against predetermined criteria. The method varies the selected architecture parameters upon failure to meet criteria until evaluation of the selected architecture parameters meets the criteria. The method selects a number of datapath clusters to avoid too many input/output ports in data registers.

Method for design of programmable data processors

A data processing system with a microprocessor. The microprocessor has an instruction execution pipeline including fetch and decode stages and several functional execution units. Fetch packets contain a plurality of instruction words. Execute packets include a plurality of instruction words that can be executed in parallel by two or more execution units. An execution packet can span two or more fetch packets. A predetermined bit in each instruction marks whether the next instruction is executed in parallel with the current instruction. Instructions in an execute packet are dispatched to appropriate functional execution units based on instruction type. Upon a branch into an execute packet instructions at memory addresses before the branch location are not executed in parallel with instructions following the branch location.

Very long instruction word microprocessor with execution packet spanning two or more fetch packets with pre-dispatch instruction selection from two latches according to instruction bit

A subpipelined translation embodiment provides binary compatibility between current an future generations of DSPs. When retrieved from memory an entire fetch packet is assigned an operating mode (base instruction set or migrant instruction set) according to the current execution mode. The fetch packets from the instruction memory are parsed into execute packets and sorted by execution unit (dispatched) in a datapath shared by both execution modes (base and migrant). The two execution modes have separate control logic. Instructions from the dispatch datapath are decoded by either base architecture decode logic or the migrant architecture decode logic, depending on the execution mode bound to the parent fetch packet. Code processed by the migrant and base decode pipelines produces machine words that are selected by a multiplexer. The multiplexer is controlled by the operating mode bound to the fetch packet that produced the machine word. The selected machine word controls a global register file, which supplies operands to all hardware execution units and accepts results of all hardware execution units.

Sub-pipelined and pipelined execution in a VLIW

A general purpose microprocessor architecture enabling more efficient computations of a type in which Boolean operations and arithmetic operations conditioned on the results of the Boolean operations are interleaved. The microprocessor is provided with a plurality of general purpose registers (&ldquo;GPRs&rdquo; )and an arithmetic logic unit (&ldquo;ALU&rdquo; ), capable of performing arithmetic operations and Boolean operations. The ALU has a first input ( ) and a second input ( ), and an output ( ), the first and second inputs receiving values stored in the GPRs. The output stores the results of the arithmetic logic unit operations in the GPRs. At least one of the GPRs is capable of receiving directly from the ALU a result of a Boolean operation. In one embodiment, at least one of the GPRs (PN)capable of receiving directly from the ALU a result of a Boolean operation is configured so as to cause the conditioning of an arithmetic operation of the ALU based on the value stored in the GPR.

Processor with conditional execution of every instruction

A microprocessor, comprising a first set of functional units capable of performing parallel data operations, a second set of functional units capable of performing parallel data operations, and a data interconnection path connecting the first and second functional units.

Microprocessor

A microprocessor 1 has an instruction fetch/decode unit 10a-c, a plurality of execution units, including an arithmetic and load/store unit D1, a multiplier M1, an ALU/shifter unit S1, an arithmetic logic unit ('ALU') L1, a shared multiport register file 20a from which data are read and to which data are written, and a memory 22. These units form an instruction execution pipeline that operates without interlocks so that nestable delayed branch instructions are provided. The control circuitry for the instruction execution pipeline is operable to begin processing a second branch instruction having a second target address on a pipeline phase immediately after beginning processing of a first branch instruction having a first target address. Furthermore, the control circuitry has no interlock or delay circuitry to condition processing of the second branch instruction based on processing of the first branch instruction, therefore the program counter circuitry receives the second target address on a pipeline phase immediately after receiving the first target address regardless of whether the first branch is taken or not. Thus, one instruction may be executed from the first target branch address and then the execution sequence can be preempted to the second target address.

Microprocessor with a nestable delayed branch instruction without branch

A general purpose microprocessor architecture enabling more efficient computations of a type in which Boolean operations and arithmetic operations conditioned on the results of the Boolean operations are interleaved. The microprocessor is provided with a plurality of general purpose registers ('GPRs' 102) and an arithmetic logic unit ('ALU' 104), capable of performing arithmetic operations and comparison operations. The ALU has a first input (108) and a second input (110), and an output (112), the first and second inputs receiving values stored in the GPRs. The output stores the results of the arithmetic logic unit operations in the GPRs. At least one of the GPRs is capable of receiving directly from the ALU a result of a Boolean operation. In one embodiment, at least one of the GPRs (PN) capable of receiving directly from the ALU a result of a Boolean operation is configured so as to cause the conditioning of an arithmetic operation of the ALU based on the value stored in the GPR.

Manipulation of boolean values and conditional operation in a

A method for compression of digital data in a computer having a processor and a memory, wherein a group of consecutive bits having the same binary value is represented by a result number corresponding to the number of the consecutive bits. The method involves the following steps. A block of digital data to be compressed is provided. A bit detect selection parameter determines a bit value to be counted for counting consecutive bits. The processor is instructed to count from a first end of the block of digital data toward a second end of the block of digital data the number of consecutive bits having the bit value determined by the bit detect selection parameter. The number of bits so counted is stored, and the bit detect selection parameter is toggled. The processor is then instructed to count from the last bit counted toward the second end of the block of digital data the number of bits having the bit value determined by the current bit detect selection parameter. The foregoing steps are repeated, with successive numbers being stored representing the number of bits counted in successive counts as successive stored result numbers, until all bits in the block of digital data are counted.

Method and apparatus for selectively counting consecutive bits

A data processing device comprising comprising a memory having a plurality of addressable memory locations, a processor circuit, an input register operative to hold input data, an output register operative to hold output data, and a direct memory access (DMA) circuit operative to receive input data from the input register for storing the input data in a first memory location and to concurrently send output data from a second memory location to said output register. Other devices, systems and methods are also disclosed.

Data processing device having split-mode DMA channel

A data processing device comprising a clock generator for producing pulses establishing instruction cycles, a memory accessible by assertion of addresses, an arithmetic logic unit connected to the memory, operative to perform an arithmetic operation on data received by the arithmetic unit. An instruction decode and control unit connected to the memory, having an instruction register operative to hold a program instruction, is operative to decode a program instruction providing control signals to control the operations of the data processing device and to initiate a interrupt sequence responsive to an instruction code having a interrupt instruction. A program sequencer circuit connected to the memory, having a program register operative to hold a program counter corresponding to a program address is operative to access the memory with the program register to obtain the program instruction corresponding to the program address. A interrupt handler unit, connected to the instruction decode and control unit and the memory, having a hold register operative to store the program register, responsive to the control signals from the instruction decode and control unit to generate an intermediate address to access an interrupt counter from the memory, is operative to store the program register into the hold register and replace the program register with the interrupt counter wherein the interrupt counter corresponds to an address for accessing an interrupt instruction to execute an interrupt routine stored in the memory. Other devices, systems and methods are also disclosed.

Processing devices with look-ahead instruction systems and methods

A data processing device comprising a storage circuit accessible by assertion of addresses, an arithmetic logic unit connected to the storage circuit operative to perform an arithmetic operation on data received by the arithmetic unit. Further included is an address register for storing an initial address word indicative of a storage circuit address. An instruction decode and control unit, connected to the storage circuit and having an instruction register operative to hold a program instruction is operative to decode the program instruction into control signals to control the operations of the data processing device and location codes to control data transfers according to predetermined sections of the program instruction wherein at least one of the sections includes a location section selecting the address register and a displacement section containing address data. Further included is an address generating unit connected to the storage circuit, the instruction register, and the address register responsive to the control signals from the instruction decode and control unit combining the initial address word from the address register and the address data from the displacement section to generate a storage circuit address. Other devices, systems and methods are also disclosed.

Processing devices with improved addressing capabilities, systems and

A data processing device comprising a clock generator for producing pulses establishing instruction cycles, a memory accessible by assertion of addresses, an instruction decode and control unit, having an instruction register operative to hold a program instruction, operative to decode a program instruction providing control signals according to a pipeline organization to control the operations of the data processing device within each instruction cycle and to initiate a block sequence responsive to an instruction code representing a block instruction. A program sequencer circuit, having a program register to hold a program count corresponding to a program address, is operative to access the memory with the contents of the program register to obtain the program instruction corresponding to the program address. Further included is an arithmetic logic unit operative to perform an arithmetic operation on data received by the arithmetic unit and to combine the contents of the program register with a data field decoded from the block instruction by the instruction decode and control unit to generate a block end address, and a block handler unit, having a block start register operative to store the contents of the program register, is responsive to the control signals from the instruction decode and control unit to store the program address corresponding to the block start address to the block start register wherein the contents of the block start register correspond to a start address for a block of instructions to be executed. Other devices, systems and methods are also disclosed.

Growing community of inventors

Richmond, TX, United States of America

Laurence Ray Simar, Jr