Abstract: A processor has 64-bit operand execution pipelines, but a 32-bit branch pipeline. The branch pipeline contains registers to hold the lower 32-bit sub-addresses of 64-bit target and sequential addresses for conditional branches which have been predicted but not yet resolved. A shared register contains the upper 32 address bits for the target and sequential sub-addresses. All 32-bit target and sequential sub-address registers in the branch pipeline share the single 32-bit shared register holding the upper 32 address bits. The branch pipeline can only process instructions with the same upper 32 address bits, which define a 4 Giga-Byte super-page. When an instruction references an address in a different 4 Giga-Byte super-page, then the branch pipeline stalls until all other branch instructions have completed. The new super-page's upper 32 address bits are then loaded into the shared register. A super-page crossing is detected by a carry out of bit 32 in the 32-bit target or sequential address adders.

Abstract: A pipelined processor is modified to efficiently process repeated string instructions. A repeated string instruction repeats an iteration a number of times determined by a counter variable stored in a register file. Each iteration includes at least three pipeline flows to perform a load, store, or compare of a character in the string, and to decrement the counter variable. When the last flow of an iteration reaches the execute stage near the end of the pipeline, the current value of the counter variable is compared to the maximum number of iterations which may be present in the pipeline at one time. When the counter variable is equal to the maximum number of iterations, the execute stage signals the decode stage to stop dispatching iterations. The iterations in the pipeline are completed, providing the proper number of iterations.

Abstract: A display interface device 20 is provided which includes inputs for receiving video data words, the video data words including control codes for controlling the output format of a display, and a video clock signal received from an associated video controller. A first-in/first-out memory 30 is also provided with a video data word clocked into memory 30 by the first clock and clocked out of memory 30 by a second clock generated from a clock received from an associated graphics controller.

Abstract: MPEG compressed video data is decompressed in a computer system by sharing computational decompression tasks between the computer system host microprocessor, the graphics accelerator, and a dedicated MPEG processor (video decoder engine) in order to make best use of resources in the computer system. Thus the dedicated MPEG processor is of minimum capability and hence advantageously minimum cost. The host microprocessor is used to decompress the MPEG upper data layers. The more powerful the host microprocessor, the more upper data layers it decompresses. The remainder of the decompression (lower data layers) is performed by the MPEG dedicated processor and/or the graphics accelerator. The video decoder engine is a fast hardwired processor. It has a graceful degradation capability to allow dropping of occasional video frames without displaying any part of a dropped video frame. The video decoder engine has a three stage pipeline structure to minimize circuitry and speed up operation.

Abstract: A dual-instruction set central processing unit (CPU) is capable of executing instructions from a reduced instruction set computer (RISC) instruction set and from a complex instruction set computer (CISC) instruction set. Data and address information may be transferred from a CISC program to a RISC program running on the CPU by using shared registers. The architecturally-defined registers in the CISC instruction set are merged or folded into some of the architecturally-defined registers in the RISC architecture so that these merged registers are shared by the two instructions sets. In particular, the flags or condition code registers defined by each architecture are merged together so that CISC instructions and RISC instructions will implicitly update the same merged flags register when performing computational instructions.

Abstract: A graphics system includes a graphics controller for rendering polygons with a minimum number of steps and registers. A host processor generates a display list that includes only the values necessary for rendering a primitive. The graphics controller includes a register file for receiving the display list either directly from the host processor or from system memory in which the host processor stored the display list. The graphics controller also includes logic to decode operational codes to ascertain which values from the register file must be used for rendering a primitive and which values can be skipped. Only the necessary values are transmitted to polygon and texture engines also included within the graphics processor.

Abstract: A mixed-modulo address generation unit has several inputs. The unit effectively adds together a subset of these inputs in a reduced modulus while simultaneously adding other inputs in a full modulus to the partial sum of reduced-modulus inputs. The subset of inputs receives reduced-width address components such as 16-bit address components which are effectively added together in modulo 64K. The other inputs receive full-width address components such as 32-bit components which are added in the full modulus, 4G. Reduced-width components are zero-extended to 32 bits before input to a standard 32-bit adder. A 16-bit carry generator also receives the reduced-width components and generates the carries out of the 16th bit position. When one or more carries is detected, a correction term is subtracted from the initial sum which is recirculated to the adder's input in a subsequent step. The correction term is the number of carries out of the 16th bit position multiplied by 64K.

Abstract: The dispatch unit of a superscalar processor checks for register dependencies among instructions to be issued together as a group. The first instruction's destination register is compared to the following instructions' sources, but the destinations of following instructions are not checked with the first instruction's destination. Instead, instructions with destination-destination dependencies are dispatched together as a group. These instructions flow down the pipelines. At the end of the pipelines the destinations are compared. If the destinations match then the results are merged together and written to the register. When instructions write to only a portion of the register, merging ensures that the correct portions of the register are written by the appropriate instructions in the group. Thus older code which performs partial-register writes can benefit from superscalar processing by dispatching the instructions together as a group and then merging the writes together at the end of the pipelines.

Abstract: An LCD controller for use e.g. in a portable computer provides gray scale shading for both monochromatic and color displays using frame rate control modulation for intensity shading for each pixel. The gray scale shading process and circuit do not require any memory for storing phase tiling matrices or frame modulation pattern sequences; both of these instead are generated in real time using a linear matrix logic structure. Use of linear matrix operations also allows generation of various phase shifts of frame modulation pattern sequences to provide a better image on the display. In addition to providing programmable 4, 8, or 16 intensity levels, the present method and apparatus provide that vertically, horizontally or diagonally adjacent pixels on the display never have the same phase in the same frame, and in addition that the pixel display drivers are uniformly loaded.

Abstract: MPEG compressed data is decompressed in a computer system by sharing computational decompression tasks between the computer system host microprocessor, the graphics accelerator, and a dedicated MPEG processor in order to make best use of resources in the computer system. Thus the dedicated MPEG processor is of minimum capability and hence advantageously minimum cost. The host microprocessor is used to decompress the MPEG upper data layers. The more powerful the host microprocessor, the more upper data layers it decompresses. The remainder of the decompression (lower data layers) is performed by the MPEG dedicated processor and/or the graphics accelerator.

Abstract: MPEG compressed data is decompressed in a computer system by sharing computational decompression tasks between the computer system host microprocessor, the graphics accelerator, and a dedicated MPEG processor in order to make best use of resources in the computer system. Thus the dedicated MPEG processor is of minimum capability and hence advantageously minimum cost. The host microprocessor is used to decompress the MPEG upper data layers. The more powerful the host microprocessor, the more upper data layers it decompresses. The remainder of the decompression (lower data layers) is performed by the MPEG dedicated processor and/or the graphics accelerator.

Abstract: A video and graphics display system compensates for video tearing caused by reading graphics data from a shared buffer faster than video data is stored into the shared buffer. The video data is arranged in video fields comprising scan lines of pixel data. A processor determines the scan line of overtake of reading graphics data from the buffer at a rate faster than storing video data of a current video field into the buffer. A generator provides at least one video scan line as an interpolation of at least one scan line of the current video field stored in the shared buffer and of at least one scan line of a previous video field stored in the shared buffer. A multiplexer receives video scan lines from the shared buffer and from the generator and provides the video scan lines from the shared buffer to a display if there is no scan line of overtake and provides the interpolated video scan lines from the generator to the display if there is a scan line of overtake.

Abstract: The frame reconstruction (FR) portion of an MPEG decompression circuit includes a horizontal interpolation element, a vertical interpolation element, and a selector (post processing) element. The horizontal and vertical interpolation elements are each digital filters averaging respectively two horizontal and two vertical adjacent pixels in an MPEG pixel block. Logic is included for constructing B, I, and P-type MPEG pictures. Also included is an error/warning handling mechanism.

Abstract: A conversion system and method is disclosed for converting between digital and analog data signals. The conversion system comprises a signal input line for each digital data signal, a reconstructor-resampler unit for each digital data signal, a combiner, a modulator, a digital-to-analog converter, and a signal output line. Each signal line input couples to the respective reconstructor-resampler unit for the digital data signal. Each reconstructor-resampler unit then couples to a combiner which couples to a modulator. The modulator couples to the digital-to-analog converter that couples to the signal output line from which an analog output signal is produced. The reconstructor-resampler comprises a sampling member coupled to the signal line input and a polynomial interpolator member coupled to the sampling member and the modulator. Also, the modulator operates at a predetermined, or fixed, frequency regardless of the sampling frequency of the digital data signal.

Abstract: Decompression of MPEG compressed audio data is performed in a computer system by the host processor in the computer system providing preprocessing data decompression and a dedicated audio decoder engine (which is a digital signal processor) performing the filtering and windowing of the host preprocessed data. The audio decoder engine includes a data path, instruction set, registers and internal program and data memory. The host performs a large portion of the audio decompression, leaving the windowing and filtering to the audio decoder engine. Thus the computationally intensive portions of the decompression are performed more efficiently. Coefficient storage in the audio decompression engine is optimized by taking advantage of the symmetries inherent in the coefficient data, both for the filter coefficients and the windowing coefficients. Double buffer input and output buffers speed the data flow between the host processor and the audio decoder engine.

Abstract: A controller for a single port (typically FIFO) memory for a computer system allocates the flow of data from a single port dynamic RAM to several data requesters such as a graphics engine, a CPU, and a display screen. One parameter is assigned to the display screen and a second parameter is assigned to the other data requester. Then each memory cycle is assigned in duty cycle fashion to the data requesters associated with one or the other of the parameters. Thus, typically the screen display which requires large amounts of data will have a relatively large parameter value associated with it such as six, while the single parameter value associated with the graphic engine and CPU will have a lower value such as two. The screen display will in response access the memory for six memory cycles consecutively, while the graphics engine and CPU will access the memory for only two memory cycles.

Abstract: An open high-speed local system bus for a microcomputer system which is decoupled from I/O and provides a consistent interface to the CPU subsystem, memory subsystem, graphics subsystem and peripheral subsystem. The local system bus supports discrete and burst transactions, pipelining in both the transactions, multiple microprocessor and distributed interrupts.

Abstract: A controller for asynchronous configurable FIFO (first-in-first-out) memory includes, in addition to two binary counters for the read pointer and write pointer, two Gray code counters for determining whether the FIFO is full or empty by a comparison of the read pointer and write pointer values expressed in Gray code. The Gray code counters avoid the problem of asynchronicity of read and write signals. The Gray code counters determine if the FIFO is full or empty depending on whether the pointer values match (indicating empty) or differ in accordance with particular Gray code patterns (indicating full). The Gray code counters each have an extra bit which allows determination of the full or empty condition from a straightforward comparison of the read pointer and write pointer values, while the Gray code eliminates the problem of multiple bit transition providing an incorrect indication of the pointer location.

Abstract: In a sequential-access memory device having storage registers located at consecutively addressable rows, data are accessed from odd and even banks by enabling consecutive rows during a common read cycle. In particular, by coupling data lines separately to odd and even register rows in the memory device, data may be accessed selectably from consecutive register rows with reduced access time.

Abstract: A dynamic random access memory (DRAM) circuit operates in burst mode when a row address strobe (RAS) signal is applied while an output enable/burst enable signal is also applied thereto. During burst mode, a column address strobe (CAS) signal is toggled to access digital data from sequential column addresses within a given row.