Abstract: A VGA controller with a read-modify-write cycle implemented therein is provided. By implementing the read-modify-write cycle in hardware, and by reducing the data for such operations to a single address source, read-modify-write operations can be performed in a single cycle, as opposed to separate read and write cycles, with a consequent improvement in overall operating speed.

Abstract: Display control logic for a terminal controller with support for such features as windows and interlace. A display list processor (DLP) (20) communicates with a program memory (12) containing DLP instructions, a display memory (12) containing character codes and attributes for the display, and a font memory (13). As the DLP program executes, it causes accesses to the display memory and brings in character codes and attributes for ultimate display on the screen. These character codes and attributes, as well as information representative of the scan line are input to a video data queue (95). The queue entries are clocked out of the queue by a character clock (170) and are used to generate addresses to font memory. Bitmaps from font memory are read into a dot shifter (190). The DLP instruction set includes a DISPLAY STRING instruction which allows a portion of a scan line to be built up by specifying the length of the scan line segment and the starting address in memory.

Abstract: A video controller for a personal computing system. The controller compensates CRT video information to generate a display compatible with a flat panel device. The controller includes registers and logic circuits which compensate CRT address information. The compensated addresses are used to repeat lines of display, insert blank lines between lines of display, center a display, and force font types.

Abstract: A memory controller which can be used with an external tag RAM is disclosed. Existing index registers in the controller serve double duty as buffers for storing tag RAM data during a test mode. Input/output lines for the external tag RAM are coupled to the index registers in addition to being coupled to a comparator for comparison with an external address during normal operation. A buffer is provided so that data from the external address from the CPU can be written through these same tag RAM input/output lines in order to update the tag RAM after a miss. In order to prevent DRAMS from putting data on the memory bus during a cache RAM test, a CAS inhibit signal is provided to the DRAM state machine. Posted writes are also disabled to avoid interference with the address provided to the tag RAM.

Abstract: In a computer system having at least two processors, each processor having an associated memory, the processors being coupled to one another through an interface unit by means of a bus, hardware semaphores to regulate access to shared resources are disclosed. Each semaphore is one bit wide and can be written to obtain the desired state. When reading the semaphore, if the contents is a one, then a one is returned. If the content is zero, a zero is returned but the semaphore is automatically reset to one.

Abstract: A method and apparatus for performing a fast jump address calculation is disclosed. A field from the instruction is provided to an adder, on the assumption that it is the displacement value, without actually determining whether it is a displacement value. A fixed instruction length is also provided to the adder, on the assumption that the instruction will have that length. Finally, the current instruction address bits from the program counter are provided to the adder. These are added together to provide a jump address.

Abstract: The present invention provides a microprocessor with a special OEM mode of operation that can be used by an OEM system integrator to implement special tasks such as power management. The OEM mode provided by the present invention is designed for use by a system integrator who integrates a microprocessor into a larger system such as a personal computer. The OEM mode of operation provided by this invention adds features to the overall system; however, it is transparent to system programmers. The system integrator can use the OEM mode to configure a system so that the system has certain special power management features, which can not be accessed by programmers who use the system. The OEM mode provided by the present invention is accessed by a signal on a special I-O pin. For example, an external timer can periodically provide signals to this special pin, thereby periodically interrupting the microprocessor and putting the microprocessor in the OEM mode.

Abstract: A circuit for waking a microprocessor from a sleep mode and providing it with its microprocessor clock long enough for a refresh, direct memory access (DMA) or master cycle operation to be done by external circuitry. The clock signal is then removed from the microprocessor to put it back into the sleep mode, thereby conserving energy. A hold signal is provided to the microprocessor to cause the microprocessor outputs to be put into a tri-state, high impedance condition, and thus relinquish control of the external bus to the external refreshing circuitry.

Abstract: A printed circuit board design capable of accepting both first and second versions of an IC device. First and second IC devices (10, 20) have pins disposed along respective first and second rectangular peripheries (12a-b and 15a-b; 22a-b and 23a-b). Each pin on the first IC device has a functional counterpart pin on the second IC device. The board configuration contains pads in first and second arrays (32a-b and 35a-b; 32a, 32c, 33a-b) that correspond to the pins on the first and second IC devices. At least some of the pads (32b) of the first array do not physically coincide with pads in the second array and are located within the rectangle defined by the second array. Each non-overlapping pad in the first array is connected by a circuit board trace (40) to a respective pad in the second array such that each circuit board trace joins two pads corresponding to counterpart pins.

Abstract: A method and apparatus for emulating the intermediate 16-bit truncation of the address in the 8086 architecture using a 32-bit adder. The preferred embodiment of the invention adds the displacement, base address, and segment base values in a three-port carry save adder. The displacement value and base address are also added together in a two-port full adder. The outputs of the three-port carry save adder and two-port full adder are then compared to determine whether a carry from bit 16, if any, resulted from the addition of the displacement and base address value or the addition of the segment base value. A logic unit determines whether a carry into bit position 16 of the linear address is modified. If the carry is the result of the addition of the segment base value to the effective address, the carry is not modified. If the carry is the result of the addition of the base and displacement value, the carry is modified by forcing the bit position 16 to zero.

Abstract: An arithmetic logic unit for a microprocessor is shown and described for use in a 24-bit data path where the ALU includes three separate ALU portions, one for each byte of the data path, and three separate control signals, one for each portion of the ALU. The ALU provides a variety of arithmetic and logic functions for application to 24-bit operands, but also includes a capability of manipulating such operands in accordance with sign extended opcodes without actually physically executing a sign extend operation within the microprocessor. In this manner, the ALU executes the necessary logic functions to provide the same ultimate result as sign bit extension, but does not require a separate sign bit extension step within the microprocessor to convert signed byte operand into a signed word operand.

Abstract: In-circuit emulation of a processor that is mounted on a circuit board. The processor (20) is provided with isolation circuitry wherein a particular input signal regime causes all processor outputs to be disabled. The emulator cable (60) terminates in a set of contacts (62a-d, 75) configured to engage the processor pins (12a-d, 25). In the special case of a surface-mount processor, the contacts are mounted to fit over and around the processor and are spring-loaded. Provision is made via at least one of the probe contacts (75) to establish the specific input signal regime at the appropriate processor pin(s) (25) in order to isolate the processor pins from any processor output signals. This prevents any processor output signals from reaching the emulator or the rest of the board logic, thereby allowing the emulator to operate.

Abstract: A video display controller capable of providing video control information for either a flat panel or a CRT display. The controller includes a plurality of main circuits, alternate circuits, select circuits, and a circuit for identifying the display device used in the system. Each main circuit receives information from the processor and generates main video information compatible with a CRT. Each alternate circuit receives information from the processor and generates alternate video information compatible with a flat panel display. Each select circuit receives main and alternate video information and outputs main video information when the display device is a CRT display and the alternate video information when the display device is a flat panel display. In one embodiment, the alternate circuits are programmable registers. In another embodiment, tables are used to program the alternate registers to provide compatibility for a number of possible display devices.

Abstract: A secondary cache control system for a computer system is disclosed. The system is utilized advantageously to reduce the cost of the SRAM while not degrading the overall performance of the CPU associated with the computer. The system latches the data from the CPU until the CPU hits a "dead time". When this dead time occurs, the data is written into the SRAM. By writing to the SRAM at this time the performance of the computer system is not degraded and the cost of the SRAM is significantly reduced.

Abstract: An apparatus and method for operating a system component in a microprocessor system. The component is operated by a component controller which runs off a clock having a frequency different than the system clock. The controller is synchronized with the system clock at the conclusion of a component access cycle. The state machine of the controller can thus operate independently of the system clock and timing options implemented by the controller need not have an even number of states.

Abstract: Two methods and apparatus for reducing power consumption in battery powered computers are disclosed. The first places the computer in a sleep mode whenever a certain data input function is called. The second applies statistical analysis to calls to another data input function. By measuring the number of times the computer has tried to read data from the keyboard over the past predefined period, the variance between the high and low number of calls over the present and preceding time periods, and whether the number of times the computer has tried to read data has both exceeded the present limit and remained within the preset variance limit for a predefined minimum time, the desirability of activating a sleep mode for the computer can be determined.

Abstract: A controller for providing O to N gray scales at a monochrome display. The monochrome display is of the type having an array of pixels energized by a display voltage over time to generate the gray scales. The controller generates a baseline time and uses the baseline time to provide gray scales at the display. In particular, each pixel is energized at least the baseline time for any gray scale above level 0 to reduce flicker in the display. In one embodiment, the baseline time corresponds to a point on the intensity response curve for the display at which the display exhibits a linear intensity response for a given display voltage versus time. In one embodiment, the baseline time is used to generate pixel on/off data to provide gray scales at the display. In yet another embodiment, the baseline time information is used to generate weighted clock information to provide gray scales at the display.

Abstract: A single semiconductor chip containing both I/O bus controller and DRAM controller functions. A single pin on the chip is used to provide both a zero wait state input to the I/O bus controller and to provide a local bus access (LBA) signal for inhibiting both the I/O bus controller and the DRAM controller when an external device is doing an I/O or memory operation on the local bus. Logic isprovided to produce an inhibit signal to the I/O bus controller in response to the LBA signal. Another logic circuit is provided to inhibit the DRAM controller in response to the LBA signal only when there is a memory cycle signal from the microprocessor. The use of the single pin is possible since the zero wait state isgnal will only appear during the latter part of an I/O or memory cycle, which is mutually exclusive with the start of an I/O or memory cycle, which is the only time the LBA signal will appear.

Abstract: A memory controller which can map EMS addresses into the DRAM behind video RAM addresses or other reserved areas of memory. A single chip has both a DRAM decoder and an EMS decoder operating in parallel. A DRAM decoder examines received addresses and provides an enable signal to a DRAM timing circuit if the address is within the DRAM range and not for a reserved group of addresses. A separate EMS decoder provides a translated address when a received address is within an EMS window. The EMS decoder also provides an EMS timing signal to the DRAM timing circuit.

Abstract: A method and apparatus for scaling interlaced video data allows the video data to be scaled to any desired size, either larger or smaller than the original image. The method considers each line of input data in each field sequentially, and determines whether the line is to be saved and whether the data lines in each successive field will also be saved, discarded, or duplicated, saving space in the output buffer for the saved and duplicated data lines to come as necessary. The method can accommodate interlaced data with N field and a scaling factor M.