摘要:
A method and apparatus is provided for enabling the transformation of a domino circuit to a static circuit without requiring the re-design of the chip or integrated circuit mask set. The domino circuit masks may be designed to include additional unconnected devices as appropriate which may be added or connected into the circuit after chip design release by changing only interconnection masks. Spare devices can be added and selectively used to make a domino circuit metal-mask programmable into a logically equivalent static circuit. In a first exemplary method, extra devices are added to, and/or existing devices are re-wired in the domino circuitry to make a complementary equivalent static gate. In a second exemplary methodology, the domino circuit is converted into a pseudo-NMOS circuit using devices already available in the circuit and modifying the circuit connections thereto.
摘要:
A reduced-power integrated circuit includes a circuit data input, a circuit data output, and at least one row of dynamic logic. The row of dynamic logic includes a row clock input, a row data input, and a row data output coupled to the circuit data output, where a value received at the row data input is derived from the value at the circuit data input. The integrated circuit further includes a comparator that compares current and previous values at the circuit data input and a switch that selectively sets the row clock signal received at the row clock input to an inactive state and temporarily maintains the row clock signal in the inactive state in response to the comparator detecting that the current previous values of at the circuit data input are equivalent. Consequently, the row of dynamic logic does not (and need not) reevaluate the circuit data input value, and power dissipation is reduced.
摘要:
A method and device are disclosed for the reduction of the penalty associated with inserting a latch in a circuit which is utilized to implement an integrated circuit in a data-processing system. A semiconductor device is disclosed which includes a main latch circuit, a feedback latch circuit and an output terminal. The main latch circuit is capable of receiving an input data signal and an input clock signal. The main latch circuit generates a latch output signal in response to the input data and clock signals. The feedback latch circuit is capable of receiving the latch output signal from the main latch circuit and storing the latch output signal. The feedback latch circuit is capable of generating a feedback latch circuit output signal which is received by the main latch circuit to maintain the latch output signal. The output terminal of the device is coupled to the feedback latch circuit for outputting a device output signal which is equal to the feedback latch circuit output signal.
摘要:
A wafer design layout and method of producing multiple integrated chip types using a single set of masks for a wafer and then at the time the type of chip desired is known, using a few customizing steps to produce the final integrated chip is provided. In one embodiment, the wafer layout includes a plurality of groupings of components and a plurality of dicing channels separating each of the components from others of the components. After the particular type of integrated circuit chip desired is selected, the wafer may then have the final few layers processed and the chips removed using the appropriate dicing channels for the integrated circuit chip desired.
摘要:
A method and apparatus is provided for changing a self-timed circuit into a self-resetting circuit to reduce the inherent delay of the self-timed circuit by an amount of latency between the assertion of the data and the assertion of the valid signal. Circuitry is provided to enable the effective de-coupling of the self-timing operation to enable data to move through the logic circuitry without the latency associated with the reception and generation of "valid" and "complete" signals being necessary. On the "receiving" side (the circuit being set into self-resetting mode), the logic circuit does not have to wait for the reception of the "valid" signals to begin operation. On the "driving" side (the circuit sending the data), the logic circuit does not have to wait for the "completion" signal to arrive to permit a new operation to occur.
摘要:
An apparatus to remove from operation a decoupling capacitor connected to a power supply providing power to logic circuitry in an integrated circuit. One technique for doing so is to connect a fuse in series with a decoupling capacitor. The present invention amplifies current transmitting through the fuse in a positive feedback manner to force the fuse to blow sooner than would normally occur. Therefore, when the current through the decoupling capacitor is deemed unacceptable, the fuse current is increased until such a time that the fuse opens.
摘要:
A method and apparatus for reducing power dissipation within a functional unit of a microprocessor includes a power sensing circuit for sensing power dissipation of the functional unit. A low power mode identifying circuit identifies when the measured power dissipation of the functional unit exceeds a predetermined amount or value. Upon such a condition, a low power mode circuit operates the functional unit in a low power mode thereby reducing its power dissipation. Operation of the functional unit in the low power mode continues until the power dissipation reaches a safe level. The functional unit internally determines power dissipation and selectively enters a low power mode to reduce power dissipation of the functional unit. Low power mode operation of the functional unit reduces power dissipation of the functional unit.
摘要:
A Self-Timed CMOS Static Circuit Technique has been invented that provides full handshaking to the source circuits; prevention of input data loss by virtue off interlocking both internal and incoming signals; full handshaking between the circuit and sink self-timed circuitry; prevention of lost access operation information by virtue of an internal lock-out for the output data information; and plug-in compatibility for some classes of dynamic self-timed systems. The net result of the overall system is that static CMOS circuits can now be used to generate a self-timed system. This is in contrast to existing self-timed systems that rely on dynamic circuits. Thus, the qualities of the static circuitry can be preserved and utilized to their fullest advantage.
摘要:
Dynamic logic circuits with reduced evaluation time provide faster output in dynamically evaluating logic circuits by increasing the rate of change of the voltage at the junction of logic input ladders. The circuits use a cross-coupled amplifier to charge the input ladder combining node once the node begins to evaluate.
摘要:
The present invention provides a method and system for providing observability of memory address access for self-timed cache designs. A system according to the present invention for determining which memory location has been accessed in a self-timed cache comprises a content addressable memory; a secondary memory coupled to the content addressable memory, wherein the secondary memory includes at least one memory location which may be selected by the content addressable memory based upon a self-timed cache access. The system further includes a test circuitry coupled to the content addressable memory, wherein the test circuitry stores a pointer which points to a selected memory location in response to the self-timed cache access of the secondary memory.