Abstract: An arithmetic unit, for example a multiply and accumulate (MAC) unit 42, for a processing engine includes a partial product reduction tree 480. The partial product reduction tree will generate carry results and provides a final output to a final adder 470 connected to the partial production reduction tree. Unbiased rounding logic 476 is provided. A carry propagation tree is responsive to the carry results for anticipating a zero on each of N least significant bits of the final adder. When zero is anticipated on each of N least significant bits of the final adder, the carry propagation tree is operable to generate an output signal 477 which is used by the unbiased rounding stage to force the (N+1)th least significant bit of the final adder to zero.

Abstract: A system and method is provided for enabling the reuse of algorithms in multiple application frameworks with no alterations required of the algorithm once it is developed. An inverted memory allocation mechanism enables various algorithm modules to be integrated into a single application without modifying the source code of the algorithm modules. During a design phase of an application, a set of algorithm modules is linked with a calling program to form an initial software program. Each of the set of algorithm modules has a memory interface which responds to a memory allocation inquiry with memory usage requirements of an instance of the algorithm module. The calling program sends a query to the memory interface of each algorithm module to request memory usage requirements for each instance of the algorithm module. A response is then sent from the memory interface of each algorithm module identifying memory usage requirements for each instance of the algorithm module.

Abstract: A digital system (100) has two or more nodes (120, 130) and a communication channel (110, 111) for transferring a single stream of ordered data from one node to another. The communication channel (110) has a number of data links (110a-110g) for transferring a plurality of sub-streams of data in a parallel fashion in order to transfer more data than a single data link is capable of transferring. Receivers (132a-132g) each have synchronizing circuitry (200, 202) for synchronizing a byte clock and a frame pulse of each received data sub-stream to the byte clock and frame pulse of a preselected master one of the receivers such that inherent data skew is eliminated.

Abstract: A processor (100) is provided that is a programmable fixed point digital signal processor (DSP) with variable instruction length, offering both high code density and easy programming. Architecture and instruction set are optimized for low power consumption and high efficiency execution of DSP algorithms, such as for wireless telephones, as well as pure control tasks. The processor includes an instruction buffer unit (106), a program flow control unit (108), an address/data flow unit (110), a data computation unit (112), and multiple interconnecting busses. Dual multiply-accumulate blocks improve processing performance. A memory interface unit (104) provides parallel access to data and instruction memories. The instruction buffer is operable to buffer single and compound instructions pending execution thereof. A decode mechanism is configured to decode instructions from the instruction buffer. The use of compound instructions enables effective use of the bandwidth available within the processor.

Abstract: Emulation and debug circuitry is provided that can be incorporated into a variety of digital systems. A stop mode of operation is provided in which an associated processor stops processing instructions in response to a debug event. A real-time mode of operation is provided in which the processor stops processing background instructions in response to a debug event, but in which high priority interrupts are still processed. While running or suspended, the emulation circuitry can jam an instruction into the instruction register of the processor to cause processor resources to be read or written on behalf of the emulation circuitry. An embodiment of a processor core is provided that is a programmable digital signal processor (DSP) with variable instruction length, offering both high code density and easy programming. Architecture and instruction set are optimized for low power consumption and high efficiency execution of DSP algorithms, such as for wireless telephones, as well as pure control tasks.

Abstract: A digital system has a host processor 200 with a bus controller 210 and peripherals 220(0)-220(31) interconnected by an interconnect bus 230. 32 peripherals are share a common strobe line (nSTROBE[0]) on an embodiment of the interconnect bus in a first subdomain. Additional sub-domains, such as sub-domain 260, can be likewise connected to interconnect bus 230. Additional strobe lines nSTROBE(n) are used to select a particular sub-domain in response to an address presented to bus controller 210 by CPU 200. An interconnect bus transaction is synchronized in background so that a current cycle is not delayed. A first write cycle 1500 is completed as a no-wait state transaction, while immediately following second write cycle 1510 is delayed while synchronization circuit 1400 completes the synchronization of the first write cycle. nSTROBE pulse 1520 indicates first write transaction 1500 while nREADY pulse 1530 indicates the completion of a no-wait state first write transaction 1500.

Abstract: A digital system is provided with a microprocessor (100), a cache (120) and various memory and devices (140a-140n). Signals to control certain cache memory modes are provided by a physical address attribute memory (PAAM) (130). For devices present in the address space of the digital system that have different capabilities and characteristics, misuse is prevented by signaling an error or otherwise limiting the use of each device in response to attribute bits in the PAAM associated with the memory mapped address of the device. A memory management unit (110) with address translation capabilities and/or memory protection features may also be present, but is not required for operation of the PAAM.

Abstract: An emulation device is provided that has a processor core that is a programmable digital signal processor (DSP). Several blocks of memory within the emulation device can be configured to emulate blocks of memory on a target processor system. Each block of memory responds to three different memory buses and can receive up the three simultaneous memory requests. Arbitration circuitry selects the highest priority memory request for service on each cycle. Each memory block is configured to respond to a block of addresses beginning at a selected starting address. Two blocks of memory can be linked to form a single merged block of memory in which both arbitration circuits operate in lock step by masking a most significant address bit of the block of address selected for the memory block.

Abstract: A microprocessor 1 is described which includes a multi-channel serial port (MCSP) 120. MCSP 120 includes clock generation and frame sync generation circuitry 300, multi-channel selection circuitry 310, and companding circuitry 320. The clock generation and frame sync generation circuitry is configurable by means of a Serial Port Control Register SPCR, and Receive Control Register RCR, a Transmit Control Register XCR, a Sample Rate Generator Register SRGR, and Pin Control Register PCR. The multi-channel selection circuitry is configurable by means of a Multi-Channel Register MCR, a Receive Channel Enable Register RCER and a Transmit Channel Enable Register XCER. Companding circuitry 320 performs optional expansion or compression of received or transmitted data using &mgr;-LAW or A-LAW, as selected by the Receive Control Register or the Transmit Control Register.

Abstract: A VIVT (virtual index, virtual tag) cache (18) uses an interruptible hardware clean function to clean dirty entries in the cache during a context switch. A MAX counter (82) and a MIN register (84) define a range of cache locations which are dirty. During the hardware clean function, the MAX counter (82) counts downward while cache entries at the address given by the MAX counter (82) are written to main memory (16) if the entry is marked as dirty. If an interrupt occurs, the MAX counter is disabled until a subsequent clean request is issued after the interrupt is serviced.

Abstract: A processor (100) is provided that is a programmable digital signal processor (DSP) with variable instruction length. A user stack region (910) is used to pass variables to a subroutine and to hold values representative of a first portion of a program counter (1000). A system stack region (911) is used to hold values representative of a remaining portion of the program counter (1001) and to hold additional context information. The user stack region and the system stack region are managed independently so that software from a prior generation processor can be translated to run on processor (100).

Abstract: An open can-type stacked capacitor is fabricated by forming a conductive layer (30, 130) outwardly of a substantially uneven surface (12, 112). A step (50, 150) is formed in an outer surface (32, 132) of the conductive layer (30, 130). A base (72, 172, 202) of a first electrode (70, 170, 200) is formed by removing a predetermined thickness (66, 166) of at least part of the conductive layer (30, 130). The base (72, 172, 202) is made of a portion of the conductive layer (30, 130) underlying the step (50, 150) by the predetermined thickness (66, 166). A sidewall (74, 174) of the first electrode (70, 170, 200) is formed. A dielectric layer (80) is formed outwardly of the first electrode (70, 170, 200). A second electrode (82) of the capacitor is formed outwardly of the dielectric layer (80).

Abstract: An integrated circuit 210 has a power grid formed from a first set of power buses 201a and 202a on a metal interconnect level M1, a second set of power buses 203a and 204a on interconnect level M4, and a third set of power buses 205a and 206a on interconnect level M5. The set of power buses on level M4 are oriented in the same direction as the set of power buses on level M1, and both sets of buses are located coincidentally. A high power logic cell 220 is pre-defined with a set of M1-M4 power vias 221 and 222 so that logic cell 220 can be positioned in a horizontal row unconstrained by pre-positioned M1-M4 power vias. Dummy cell 230 with M1-M4 power vias is positioned as needed so as not to exceed a maximum strapping distance D1. A maximum value for distance D1 is selected based on dynamic power requirements of nearby logic cells 250a-n as determined by simulation. A method for designing and fabricating integrated circuit 210 is described.

Abstract: A data processing system having a central processing (CPU) unit and a method of operation is provided. The CPU has an instruction set architecture that is optimized for intensive numeric algorithm processing. The CPU has dual load/store units connected to dual memory target ports of a memory controller. The CPU can execute two aligned data transfers each having a length of one byte, two bytes, four bytes, or eight bytes in parallel by executing two load/store instructions. The CPU can also execute a single non-aligned data transfer having a length of four bytes or eight bytes by executing a non-aligned load/store instruction that utilizes both memory target ports. A data transfer address for each load/store instruction is formed by fetching the instruction, decoding the instruction to determine instruction type, transfer data size, and scaling selection, selectively scaling an offset provided by the instruction and combining the selectively scaled offset with a base address value.

Abstract: A multiply-accumulate (MAC) unit, having a first binary operand X, a second binary operand Y, a third binary operand, Booth recode logic for generating a plurality of partial products from said first and second operands, a Wallace tree adder for reducing the partial products and for selectively arithmetically combining the reduced partial products with said third operand, a final adder for generating a final sum, and a saturation circuitry for selectively rounding or saturating said final sum is provided. A dual MAC unit is also provided.

Abstract: A digital system is provided with an interface circuit for interconnecting two modules in different clock domains. The interface circuit can selectively respond to a request signal from a remote source and immediately de-assert a ready signal in response to either a rising edge or a falling edge of the request signal asynchronously to a local clock signal. When an internal circuit connected to the interface circuit has completed a requested operation, the interface circuit asserts the ready signal. The ready signal is de-asserted and asserted in a glitchless manner so that the remote module can respond to the ready signal asynchronously to a remote clock signal.

Abstract: Emulation and debug circuitry is provided that can be incorporated into a variety of digital systems. A stop mode of operation is provided in which an associated processor stops processing instructions in response to a debug event. A real-time mode of operation is provided in which the processor stops processing background instructions in response to a debug event, but in which high priority interrupts are still processed. Interrupts are classified and processed accordingly when the processor is stopped by a debug event. While suspended for a debug event, a frame counter keeps track of interrupt debug state if multiple interrupts occur. While running or suspended, the emulation circuitry can jam an instruction into the instruction register of the processor to cause processor resources to be read or written on behalf of the emulation circuitry. Read/write transactions are qualified by an expected frame count to maintain correspondence between test host software and multiple debug/interrupt events.

Abstract: Emulation and debug circuitry is provided that can be incorporated into a variety of digital systems. A stop mode of operation is provided in which an associated processor stops processing instructions in response to a debug event. A real-time mode of operation is provided in which the processor stops processing background instructions in response to a debug event, but in which high priority interrupts are still processed. Interrupts are classified and processed accordingly when the processor is stopped by a debug event. While suspended for a debug event, a frame counter keeps track of interrupt debug state if multiple interrupts occur. While running or suspended, the emulation circuitry can jam an instruction into the instruction register of the processor to cause processor resources to be read or written on behalf of the emulation circuitry. Read/write transactions are qualified by an expected frame count to maintain correspondence between test host software and multiple debug/interrupt events.

Abstract: Emulation and debug circuitry is provided that can be incorporated into a variety of digital systems. A stop mode of operation is provided in which an associated processor stops processing instructions in response to a debug event. A real-time mode of operation is provided in which the processor stops processing background instructions in response to a debug event, but in which high priority interrupts are still processed. Interrupts are classified and processed accordingly when the processor is stopped by a debug event. While suspended for a debug event, a frame counter keeps track of interrupt debug state if multiple interrupts occur. While running or suspended, the emulation circuitry can jam an instruction into the instruction register of the processor to cause processor resources to be read or written on behalf of the emulation circuitry. Read/write transactions are qualified by an expected frame count to maintain correspondence between test host software and multiple debug/interrupt events.

Abstract: A processing engine 10 provides computation of an output vector as a linear combination of N input vectors with N coefficients in an efficient manner. The processing engine includes a coefficient register 940 for holding a representation of each of N coefficients of a first input vector. A test unit 950 is provided for testing selected parts (e.g. bits) of the coefficient register for respective coefficient representations. An arithmetic unit 970 computes respective coordinates of an output vector by selective addition/subtraction of coordinates of a second input vector dependent on results of the coefficient representation tests. Power consumption can be kept low due to the use of a coefficient test operation in parallel with an ALU operation.