US20020097069A1 - Domino logic with self-timed precharge - Google Patents
Domino logic with self-timed precharge Download PDFInfo
- Publication number
- US20020097069A1 US20020097069A1 US09/769,075 US76907501A US2002097069A1 US 20020097069 A1 US20020097069 A1 US 20020097069A1 US 76907501 A US76907501 A US 76907501A US 2002097069 A1 US2002097069 A1 US 2002097069A1
- Authority
- US
- United States
- Prior art keywords
- logic
- precharge
- state
- stage
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000011156 evaluation Methods 0.000 claims abstract description 30
- 230000003111 delayed effect Effects 0.000 claims description 11
- 238000000034 method Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims 3
- 230000008030 elimination Effects 0.000 abstract description 2
- 238000003379 elimination reaction Methods 0.000 abstract description 2
- 230000007704 transition Effects 0.000 description 11
- 239000003990 capacitor Substances 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000000295 complement effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03K—PULSE TECHNIQUE
- H03K19/00—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits
- H03K19/02—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components
- H03K19/08—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices
- H03K19/094—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices using field-effect transistors
- H03K19/096—Synchronous circuits, i.e. using clock signals
- H03K19/0966—Self-timed logic
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03K—PULSE TECHNIQUE
- H03K19/00—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits
- H03K19/02—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components
- H03K19/08—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices
- H03K19/094—Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices using field-effect transistors
- H03K19/096—Synchronous circuits, i.e. using clock signals
- H03K19/0963—Synchronous circuits, i.e. using clock signals using transistors of complementary type
Definitions
- This invention relates to the field of electronic circuit design, and in particular to the design of logic circuits for use in a domino-chain of logic devices.
- domino logic propagates logic values from one stage to the next; a first stage propagates a logic result to a second stage, which propagates the result to a third stage, and so on.
- a clock signal is used to preset all of the stages during a first phase of the clock, and to enable the stages to change state in a second phase of the clock, depending upon the input signals and logic function of each stage.
- Each stage is configured to provide an inactive state when it is precharged, such that this stage can have no effect on the next stage while it is in the inactive state. The state remains inactive until the stage is enabled and the application of the input to the gate's logic function results in a change of state to an active state; only then will the next stage potentially change state, depending upon the application of this active state to the next stage's logic function.
- FIG. 1 illustrates an example domino CMOS circuit 100 comprising a series of domino stages 110 a - c.
- each stage includes a P-channel transistor 120 p and an N-channel transistor 120 n in series with its logic function 130 a - c, each transistor 120 p, 120 n being gated by a common clock, or precharge, signal 141 , such that only one of the transistors is conducting at any time, thereby precluding the flow of DC current, for low power consumption.
- Both transistors 120 p and 120 n are required, to prevent the flow of DC current when the precharge transistor 120 p conducts and the logic function 130 also conducts.
- the logic function 130 may be any combination of transistors, but generally includes transistors of one type, either p-channel or n-channel. Because n-channel devices are generally faster than p-channel devices of equal size, the logic function 130 in high-speed designs generally include only n-channel devices.
- Logic function 130 b is illustrated as a two-input OR gate, implemented as a NOR combination of n-channel devices 131 , 132 , and an inverter 135 .
- the clock is structured to precharge the logic function 130 via the p-channel device when it is at a logic-0 level, and to ‘evaluate’ the logic function when it is at the logic-1 level.
- each logic function 130 when the clock is in the precharge state (logic-0), the transistors 120 p conduct, and the output of each logic function 130 provides a logic-0 output, via, for example, a corresponding inverter 135 in each logic function 130 . Because the logic function 130 includes only n-channel devices, the logic-0 output from one stage cannot alter the logic state of a subsequent stage. To avoid noise-induced transients, a weak-latch (not shown) is often used to hold the output at a logic-0 state until it is actively driven to a logic-1 state by a discharge through the n-channel devices.
- n-channel transistor 120 n would be used to provide a ‘precharge’ to logic-0, and the preceding stage would be configured to provide a logic-1 state as the precharge state.
- Alternating n-channel and p-channel stages may be employed to eliminate the need for the inverter 135 in each logic function 130 .
- Other configurations, including p-channel and n-channel devices within a logic function 130 (with appropriate precharge states on each input) are also feasible.
- the speed of a domino stage is determined by the delay of the logic function, plus the delay through the evaluation transistor, 120 p or 120 n.
- the logic functions 130 in a typical design such as an adder or multiplier, include only two or three inputs, and thereby a maximum stack size of two or three transistors in series.
- the evaluation transistor 120 can amount to a third or a quarter of the propagation delay of each stage. A large size evaluation transistor will reduce the delay through the transistor, but at the cost of circuit area, and increased loading on the clock circuit (and thereby increased switching power consumption).
- FIG. 1 illustrates an example circuit comprising domino logic stages, as is known in the art.
- FIG. 2 illustrates an example circuit comprising a modified domino logic stage in accordance with this invention.
- FIG. 3 illustrates an example circuit comprising an alternative precharge enabling structure for controlling the precharge of a modified domino logic stage in accordance with this invention.
- FIG. 4 illustrates a preferred embodiment of a domino logic structure in accordance with this invention.
- this invention is presented herein using the paradigm of domino stages having p-channel precharge and n-channel discharge structures.
- alternative structures including combinations of p-channel and n-channel precharge structures, and corresponding n-channel and p-channel discharge structures, may also be used.
- the principles of this invention are applicable to any logic stages that conventionally use a combination of precharge devices and evaluation devices to avoid a DC current path from power to ground during the precharge phase of each precharge-evaluate cycle.
- FIG. 2 illustrates a logic network 200 that includes a modified domino logic stage 210 in accordance with this invention.
- the logic stage 210 includes the conventional precharge device 120 p, but does not include the conventional evaluation device 120 n that is illustrated in the first stage 110 a.
- the evaluation device 120 n is used in a conventional precharge logic stage 110 a to assure that a DC current path between power and ground cannot occur when the precharge device 120 p conducts.
- the DC current path between power and ground is prevented by assuring that the precharge device 120 p does not conduct until the logic function 130 is placed into a non-conducting state.
- the inactive state of an output is configured to place the devices in the logic function block 130 of a subsequent logic stage 110 b into a non-conducting state, so that the precharged state of the subsequent stage 110 b is retained until it is actively discharged via the logic function 130 b and the evaluation device ( 120 n in FIG. 1), after the discharge device 120 p is placed in a non-conducting state.
- This invention is based on the observation that when the inputs to a domino logic block are placed in the inactive state, current cannot flow through the logic function 130 . Therefore, by delaying the application of the precharge signal until the inputs are in the inactive state, a DC current path from power to ground can be avoided, without the use of an evaluation device 120 n that also assures that a DC current path from power to ground does not occur when the stage 110 is being precharged.
- a precharge enabler 240 is configured to apply the precharge signal 241 b to the precharge device 120 p of logic stage 210 .
- the precharge enabler 240 receives the inputs to the logic stage 210 , and the original precharge signal 141 , and provides the precharge signal 241 b only when each of the inputs to the logic function 130 b is in the inactive state.
- the precharge enabler 240 b does not assert the precharge signal 241 b until each of the inputs to the logic function 130 b is at the logic-0 state, and the precharge signal 141 is also asserted. Because each of the inputs to the logic function 130 b is in the inactive state, and therefore the logic function 130 b is in a non-conducting state, DC current cannot flow between power and ground when the precharge signal 241 is applied.
- a precharge enabler 240 c is provided to delay the assertion of a precharge signal 241 c until all of the inputs to a subsequent stage are in the inactive state, and so on.
- the precharge enabler 240 c may be configured to delay either the original precharge signal 141 or the delayed precharge signal 241 b, because the delayed precharge signal 241 c need not be asserted until after the assertion of the delayed precharge signal 241 b, and a cascaded structure as illustrated in FIG. 2 reduces the load on the original precharge signal 141 .
- FIG. 3 illustrates an alternative embodiment of a logic network 300 that allows the use of a logic stage 210 that does not include an evaluation device, and does not add an additional load to the logic signals.
- the purpose of the precharge enabler 240 b of FIG. 2 is to delay the assertion of the precharge signal to the logic stage 210 until each of the inputs to the logic stage 210 is in the inactive state.
- a precharge delay 340 b is used to provide this fixed-delay assertion of the precharge signal 341 b to the logic stage 210 .
- the logic function 130 b will be in the non-conductive state, and the assertion of the delated precharge signal 341 b will not result in a DC current path from power to ground.
- a precharge delay 340 c is used to provide a delayed precharge signal 341 c for a subsequent stage, wherein the fixed-delay of this delayed precharge signal 341 c is based on the maximum delay time for placing the outputs of the prior stages into an inactive state.
- the precharge signal to subsequent stages will be similarly delayed, based on the delay associated with placing the outputs of the prior stages into an inactive state.
- the input to the logic stage 210 includes an input from other precharge logic blocks (not shown). These logic blocks are effectively operated in parallel with the logic stage 110 a.
- the precharge delay block 340 b is configured to delay the assertion of the precharge signal 341 b to the precharge device 120 p of the logic stage 210 based on the maximum precharge-delay of each of the logic blocks that provide an input to the logic stage 210 .
- other precharge logic blocks typically operate in parallel to the logic stage 210 to provide inputs to a subsequent stage (not shown).
- the precharge delay block 340 c is configured to delay the assertion of the precharge signal 341 c based on the maximum precharge-delay of the blocks that provide inputs to the subsequent stage. As such, the fixed delay of each of the precharge delay blocks 340 b, 340 c may differ.
- the total duration of the evaluation period is determined based on the maximum delay to bring an output of the last stage of the logic structure to an active state, based on an active state of an input to the first stage of the logic structure, measured from the time of assertion of the evaluation signal (or, de-assertion of the precharge signal).
- This total duration will include, in a conventional domino logic structure, the additional delay associated with each evaluation device of each stage.
- the evaluation device is only used in the first stage, and therefore, the total duration of the evaluation period is substantially reduced.
- a reduction of (N ⁇ 1)*(the delay of the evaluation device) can, ideally, be achieved via the application of the principles of this invention.
- FIG. 4 illustrates a preferred embodiment of this invention.
- the precharge delay is provided by a dummy timing cell 440 , and precharge enabling logic 448 .
- the dummy timing cell 440 includes a logic structure 442 that is similar in structure to the logic blocks 130 in the logic stages 410 .
- the actual delay of the timing cell 440 is preset to correspond to the maximum delay associated with the precharge propagation delay in a prior logic stage, as noted above. This preset delay may be embodied by any of a variety of techniques, common in the art, and symbolically illustrated in FIG. 4 as an adjustable load capacitance on the logic structure 442 .
- the capacitor 444 When the original precharge signal 141 is in the logic-1 state, the capacitor 444 is discharged, and the inverter 446 provides a logic-1 output. At the same time, an inversion 141 ′ of the precharge signal 141 is in the logic-0 state, and the precharge signal 441 is a logic-1, and places the precharge device 120 p into a non-conductive state.
- the complementary precharge signal 141 ′ transitions to a logic-1 state, and places the enabling device 448 n into a conducting state.
- the asserted precharge signal 141 places the precharge device 442 p into a conducting state, and begins to charge the capacitor 444 .
- the inverter 446 asserts a logic-0 as the precharge signal 441 to precharge device 120 p in the logic stage 410 .
- the capacitor 444 is sized so that the delay of the transition of the precharge signal 141 to the assertion of the precharge signal 441 is sufficient long to assure that all of the logic signals from the prior stage have reached the inactive state, as discussed above.
- the enabling device 448 n When the original precharge signal 141 transitions to a logic-1 state, to de-assert precharge, the enabling device 448 n enters the non-conductive state, and the capacitor 444 is discharged through the series of n-channel devices in the logic structure 442 .
- the enabling device 448 p brings the delayed precharge signal 441 to a de-asserted, logic-1 state.
- the transition of the original precharge signal 141 corresponds to a transition into the evaluation phase, and the first stage ( 110 a of FIG. 3) of the logic structure begins the domino-propagation of the logic values produced in response to the first stage input signals.
- FIG. 4 also illustrates an optional device 415 that is used to bring the output of the logic stage 410 to an inactive state as soon as the domino logic structure enters the precharge phase.
- the timing cells 440 can be configured to correspond to this reduced delay, thereby providing an optimization of the precharge phase.
- the duration of the precharge phase is configured to assure that the output of all the logic functions 130 do, in fact, enter the inactive state before the corresponding pre-charge signal is de-asserted, to assure that the logic output 441 does not return to the active state when the device 415 is placed in the non-conductive state.
- the device 415 is optional, because although it decreases the required precharge duration, it increases the power consumption, because the output of the logic function 130 may be actively driven high (via the inverter 135 of FIG. 1, for example), and the enabling of the device 415 will produce a DC path to ground. Assuming that each stage in the logic structure is configured with a corresponding device 415 , however, the duration of this DC current flow is merely the transition time of the logic function output device ( 135 of FIG. 1).
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Logic Circuits (AREA)
Abstract
Description
- 1. Field of the Invention
- This invention relates to the field of electronic circuit design, and in particular to the design of logic circuits for use in a domino-chain of logic devices.
- 2. Description of Related Art
- The use of domino logic is well known in the art. As the name implies, a domino logic circuit propagates logic values from one stage to the next; a first stage propagates a logic result to a second stage, which propagates the result to a third stage, and so on. Typically, a clock signal is used to preset all of the stages during a first phase of the clock, and to enable the stages to change state in a second phase of the clock, depending upon the input signals and logic function of each stage. Each stage is configured to provide an inactive state when it is precharged, such that this stage can have no effect on the next stage while it is in the inactive state. The state remains inactive until the stage is enabled and the application of the input to the gate's logic function results in a change of state to an active state; only then will the next stage potentially change state, depending upon the application of this active state to the next stage's logic function.
- FIG. 1 illustrates an example
domino CMOS circuit 100 comprising a series of domino stages 110 a-c. In this example, each stage includes a P-channel transistor 120 p and an N-channel transistor 120 n in series with itslogic function 130 a-c, each 120 p, 120 n being gated by a common clock, or precharge,transistor signal 141, such that only one of the transistors is conducting at any time, thereby precluding the flow of DC current, for low power consumption. Both 120 p and 120 n are required, to prevent the flow of DC current when thetransistors precharge transistor 120 p conducts and thelogic function 130 also conducts. - The
logic function 130 may be any combination of transistors, but generally includes transistors of one type, either p-channel or n-channel. Because n-channel devices are generally faster than p-channel devices of equal size, thelogic function 130 in high-speed designs generally include only n-channel devices.Logic function 130 b is illustrated as a two-input OR gate, implemented as a NOR combination of n- 131, 132, and anchannel devices inverter 135. In this example, the clock is structured to precharge thelogic function 130 via the p-channel device when it is at a logic-0 level, and to ‘evaluate’ the logic function when it is at the logic-1 level. - In this n-channel logic example, when the clock is in the precharge state (logic-0), the
transistors 120 p conduct, and the output of eachlogic function 130 provides a logic-0 output, via, for example, acorresponding inverter 135 in eachlogic function 130. Because thelogic function 130 includes only n-channel devices, the logic-0 output from one stage cannot alter the logic state of a subsequent stage. To avoid noise-induced transients, a weak-latch (not shown) is often used to hold the output at a logic-0 state until it is actively driven to a logic-1 state by a discharge through the n-channel devices. - When the clock transitions to the evaluate state (logic-1), the p-
channel transistors 120 p cease conduction, and the n-channel devices 120 n conduct, allowing thelogic function 130 to change from the precharge state, as determined by its input state. Note that, in an n-channel logic function 130, a transition to the active, non-precharge, state cannot occur unless an input transitions to a logic-1 state. That is, changes in state propagate sequentially through the stages 110 a-110 b-110 c, in a falling-domino-like manner. - As would be evident to one of ordinary skill in the art, if a
logic function 130 includes only p-channel devices, the n-channel transistor 120 n would be used to provide a ‘precharge’ to logic-0, and the preceding stage would be configured to provide a logic-1 state as the precharge state. Alternating n-channel and p-channel stages may be employed to eliminate the need for theinverter 135 in eachlogic function 130. Other configurations, including p-channel and n-channel devices within a logic function 130 (with appropriate precharge states on each input) are also feasible. - The speed of a domino stage is determined by the delay of the logic function, plus the delay through the evaluation transistor, 120 p or 120 n. Generally, the logic functions 130 in a typical design, such as an adder or multiplier, include only two or three inputs, and thereby a maximum stack size of two or three transistors in series. Assuming equally sized transistors, the evaluation transistor 120 can amount to a third or a quarter of the propagation delay of each stage. A large size evaluation transistor will reduce the delay through the transistor, but at the cost of circuit area, and increased loading on the clock circuit (and thereby increased switching power consumption).
- It is an object of this invention to eliminate the delay caused by the use of an evaluation transistor in a pre-charged logic stage. It is a further object of this invention to reduce the circuit area required in a pre-charged logic stage. It is a further object of this invention to reduce the load of a clock circuit in a pre-charged logic design. It is a further object of this invention to reduce the power consumption of a pre-charged logic design.
- These objects and others are achieved by controlling the precharge of a logic stage based on the precharge delay of a prior logic stage. The precharge of the logic stage does not occur until the output of the prior logic stage corresponds to the precharge logic state. Because the precharge logic state output of a preceding stage is an inactive state of a subsequent logic stage, the logic function of the subsequent logic stage is in a non-conducting state when the output of the prior logic stage is in the precharge logic state. By providing the precharge to a subsequent stage only after the output of the prior stage is in the precharge state, there can be no DC current flow during the precharge of the subsequent stage, and the need for an evaluation transistor to block the DC current flow during precharge is eliminated. The elimination of the evaluation transistor eliminates the delay introduced by the evaluation transistor in a precharge logic stage, reduces the circuit area for the logic stage, reduces the load on the clock circuit, and reduces the power consumption of each logic stage.
- The invention is explained in further detail, and by way of example, with reference to the accompanying drawings wherein:
- FIG. 1 illustrates an example circuit comprising domino logic stages, as is known in the art.
- FIG. 2 illustrates an example circuit comprising a modified domino logic stage in accordance with this invention.
- FIG. 3 illustrates an example circuit comprising an alternative precharge enabling structure for controlling the precharge of a modified domino logic stage in accordance with this invention.
- FIG. 4 illustrates a preferred embodiment of a domino logic structure in accordance with this invention.
- Throughout the drawings, the same reference numerals indicate similar or corresponding features or functions.
- For ease of reference and understanding, this invention is presented herein using the paradigm of domino stages having p-channel precharge and n-channel discharge structures. As noted above, alternative structures, including combinations of p-channel and n-channel precharge structures, and corresponding n-channel and p-channel discharge structures, may also be used. As will be evident to one of ordinary skill in the art, the principles of this invention are applicable to any logic stages that conventionally use a combination of precharge devices and evaluation devices to avoid a DC current path from power to ground during the precharge phase of each precharge-evaluate cycle.
- FIG. 2 illustrates a
logic network 200 that includes a modifieddomino logic stage 210 in accordance with this invention. As illustrated thelogic stage 210 includes theconventional precharge device 120 p, but does not include theconventional evaluation device 120 n that is illustrated in thefirst stage 110 a. As noted above, theevaluation device 120 n is used in a conventionalprecharge logic stage 110 a to assure that a DC current path between power and ground cannot occur when theprecharge device 120 p conducts. - In accordance with this invention, the DC current path between power and ground is prevented by assuring that the
precharge device 120 p does not conduct until thelogic function 130 is placed into a non-conducting state. As discussed above, in a conventionaldomino logic stage 110 a, the inactive state of an output is configured to place the devices in thelogic function block 130 of asubsequent logic stage 110 b into a non-conducting state, so that the precharged state of thesubsequent stage 110 b is retained until it is actively discharged via thelogic function 130 b and the evaluation device (120 n in FIG. 1), after thedischarge device 120 p is placed in a non-conducting state. - This invention is based on the observation that when the inputs to a domino logic block are placed in the inactive state, current cannot flow through the
logic function 130. Therefore, by delaying the application of the precharge signal until the inputs are in the inactive state, a DC current path from power to ground can be avoided, without the use of anevaluation device 120 n that also assures that a DC current path from power to ground does not occur when the stage 110 is being precharged. - As illustrated in FIG. 2, a precharge enabler 240 is configured to apply the
precharge signal 241 b to theprecharge device 120 p oflogic stage 210. The precharge enabler 240 receives the inputs to thelogic stage 210, and theoriginal precharge signal 141, and provides theprecharge signal 241 b only when each of the inputs to thelogic function 130 b is in the inactive state. That is, in the n-channel logic function example, wherein the inactive state for an n-channel device is a logic-0, theprecharge enabler 240 b does not assert theprecharge signal 241 b until each of the inputs to thelogic function 130 b is at the logic-0 state, and theprecharge signal 141 is also asserted. Because each of the inputs to thelogic function 130 b is in the inactive state, and therefore thelogic function 130 b is in a non-conducting state, DC current cannot flow between power and ground when the precharge signal 241 is applied. - In like manner, a
precharge enabler 240 c is provided to delay the assertion of aprecharge signal 241 c until all of the inputs to a subsequent stage are in the inactive state, and so on. Theprecharge enabler 240 c may be configured to delay either theoriginal precharge signal 141 or thedelayed precharge signal 241 b, because thedelayed precharge signal 241 c need not be asserted until after the assertion of thedelayed precharge signal 241 b, and a cascaded structure as illustrated in FIG. 2 reduces the load on theoriginal precharge signal 141. - Because the example
precharge enabler 240 b receives the inputs to thesubsequent logic stage 210, theprecharge enabler 240 b adds an additional load to each of the inputs, thereby decreasing the speed of transition of the inputs, and increasing the power consumption. FIG. 3 illustrates an alternative embodiment of alogic network 300 that allows the use of alogic stage 210 that does not include an evaluation device, and does not add an additional load to the logic signals. As noted above, the purpose of theprecharge enabler 240 b of FIG. 2 is to delay the assertion of the precharge signal to thelogic stage 210 until each of the inputs to thelogic stage 210 is in the inactive state. If the maximum time delay, after the assertion of theprecharge signal 141, for the output of alogic stage 110 a to achieve the inactive state is known, then the delay in the assertion of theprecharge signal 241 b to the subsequent stage can be delayed by this maximum amount, thereby avoiding the need to monitor the input signals to thesubsequent stage 210 to determine when they achieve the inactive state. As illustrated in FIG. 3, aprecharge delay 340 b is used to provide this fixed-delay assertion of theprecharge signal 341 b to thelogic stage 210. Because the outputs of the prior stages will be in the inactive state after this fixed-delay, thelogic function 130 b will be in the non-conductive state, and the assertion of the delatedprecharge signal 341 b will not result in a DC current path from power to ground. - In like manner, a
precharge delay 340 c is used to provide a delayedprecharge signal 341 c for a subsequent stage, wherein the fixed-delay of this delayedprecharge signal 341 c is based on the maximum delay time for placing the outputs of the prior stages into an inactive state. The precharge signal to subsequent stages will be similarly delayed, based on the delay associated with placing the outputs of the prior stages into an inactive state. - Note that the input to the
logic stage 210 includes an input from other precharge logic blocks (not shown). These logic blocks are effectively operated in parallel with thelogic stage 110 a. Theprecharge delay block 340 b is configured to delay the assertion of theprecharge signal 341 b to theprecharge device 120 p of thelogic stage 210 based on the maximum precharge-delay of each of the logic blocks that provide an input to thelogic stage 210. In like manner, other precharge logic blocks typically operate in parallel to thelogic stage 210 to provide inputs to a subsequent stage (not shown). Theprecharge delay block 340 c is configured to delay the assertion of theprecharge signal 341 c based on the maximum precharge-delay of the blocks that provide inputs to the subsequent stage. As such, the fixed delay of each of the precharge delay blocks 340 b, 340 c may differ. - In a conventional domino logic structure, the total duration of the evaluation period is determined based on the maximum delay to bring an output of the last stage of the logic structure to an active state, based on an active state of an input to the first stage of the logic structure, measured from the time of assertion of the evaluation signal (or, de-assertion of the precharge signal). This total duration will include, in a conventional domino logic structure, the additional delay associated with each evaluation device of each stage. In accordance with this invention, however, the evaluation device is only used in the first stage, and therefore, the total duration of the evaluation period is substantially reduced. In an N-stage domino logic structure, a reduction of (N−1)*(the delay of the evaluation device) can, ideally, be achieved via the application of the principles of this invention.
- FIG. 4 illustrates a preferred embodiment of this invention. In this embodiment, the precharge delay is provided by a
dummy timing cell 440, and precharge enablinglogic 448. Thedummy timing cell 440 includes alogic structure 442 that is similar in structure to the logic blocks 130 in the logic stages 410. By using a similar logic structure, the timing characteristics of thelogic function 130 that vary with environmental conditions, or fabrication conditions, will be reflected in thetiming cell 440, to provide a delay that varies with the variations of the delay of thelogic function 130. The actual delay of thetiming cell 440 is preset to correspond to the maximum delay associated with the precharge propagation delay in a prior logic stage, as noted above. This preset delay may be embodied by any of a variety of techniques, common in the art, and symbolically illustrated in FIG. 4 as an adjustable load capacitance on thelogic structure 442. - When the original
precharge signal 141 is in the logic-1 state, thecapacitor 444 is discharged, and theinverter 446 provides a logic-1 output. At the same time, aninversion 141′ of theprecharge signal 141 is in the logic-0 state, and theprecharge signal 441 is a logic-1, and places theprecharge device 120 p into a non-conductive state. - When the original
precharge signal 141 transitions to a logic-0 state, to assert precharge, the complementaryprecharge signal 141′ transitions to a logic-1 state, and places the enablingdevice 448 n into a conducting state. The assertedprecharge signal 141 places the precharge device 442 p into a conducting state, and begins to charge thecapacitor 444. When the voltage on thecapacitor 444 rises above the switching threshold of theinverter 446, theinverter 446 asserts a logic-0 as theprecharge signal 441 toprecharge device 120 p in thelogic stage 410. Thecapacitor 444 is sized so that the delay of the transition of theprecharge signal 141 to the assertion of theprecharge signal 441 is sufficient long to assure that all of the logic signals from the prior stage have reached the inactive state, as discussed above. - When the original
precharge signal 141 transitions to a logic-1 state, to de-assert precharge, the enablingdevice 448 n enters the non-conductive state, and thecapacitor 444 is discharged through the series of n-channel devices in thelogic structure 442. The enablingdevice 448 p brings the delayedprecharge signal 441 to a de-asserted, logic-1 state. The transition of the originalprecharge signal 141 corresponds to a transition into the evaluation phase, and the first stage (110 a of FIG. 3) of the logic structure begins the domino-propagation of the logic values produced in response to the first stage input signals. - FIG. 4 also illustrates an
optional device 415 that is used to bring the output of thelogic stage 410 to an inactive state as soon as the domino logic structure enters the precharge phase. This eliminates the delay of the precharge from theprecharge input 441 to the logic output 411. By providing this reduced-delay inactive output, the timingcells 440 can be configured to correspond to this reduced delay, thereby providing an optimization of the precharge phase. The duration of the precharge phase is configured to assure that the output of all the logic functions 130 do, in fact, enter the inactive state before the corresponding pre-charge signal is de-asserted, to assure that thelogic output 441 does not return to the active state when thedevice 415 is placed in the non-conductive state. - As noted above, the
device 415 is optional, because although it decreases the required precharge duration, it increases the power consumption, because the output of thelogic function 130 may be actively driven high (via theinverter 135 of FIG. 1, for example), and the enabling of thedevice 415 will produce a DC path to ground. Assuming that each stage in the logic structure is configured with acorresponding device 415, however, the duration of this DC current flow is merely the transition time of the logic function output device (135 of FIG. 1). - The foregoing merely illustrates the principles of the invention. It will thus be appreciated that those skilled in the art will be able to devise various arrangements which, although not explicitly described or shown herein, embody the principles of the invention and are thus within the spirit and scope of the following claims.
Claims (14)
Priority Applications (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US09/769,075 US6420904B1 (en) | 2001-01-25 | 2001-01-25 | Domino logic with self-timed precharge |
| KR1020027012660A KR20030003702A (en) | 2001-01-25 | 2002-01-18 | Domino logic with self-timed precharge |
| CN02800816.2A CN1582532A (en) | 2001-01-25 | 2002-01-18 | Domino logic with self-timed precharge |
| JP2002560278A JP2004523950A (en) | 2001-01-25 | 2002-01-18 | Domino logic for self-time precharge |
| PCT/IB2002/000188 WO2002060061A2 (en) | 2001-01-25 | 2002-01-18 | Domino logic with self-timed precharge |
| TW091100937A TW535355B (en) | 2001-01-25 | 2002-01-22 | Domino logic with self-timed precharge |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US09/769,075 US6420904B1 (en) | 2001-01-25 | 2001-01-25 | Domino logic with self-timed precharge |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| US6420904B1 US6420904B1 (en) | 2002-07-16 |
| US20020097069A1 true US20020097069A1 (en) | 2002-07-25 |
Family
ID=25084372
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US09/769,075 Expired - Fee Related US6420904B1 (en) | 2001-01-25 | 2001-01-25 | Domino logic with self-timed precharge |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US6420904B1 (en) |
| JP (1) | JP2004523950A (en) |
| KR (1) | KR20030003702A (en) |
| CN (1) | CN1582532A (en) |
| TW (1) | TW535355B (en) |
| WO (1) | WO2002060061A2 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102012111414A1 (en) * | 2012-11-26 | 2014-05-28 | Infineon Technologies Ag | memory circuit |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7034578B2 (en) * | 2003-04-28 | 2006-04-25 | Via Technologies, Inc. | N-domino output latch with accelerated evaluate path |
| US7187210B2 (en) * | 2003-08-13 | 2007-03-06 | Via Technologies, Inc. | P-domino register |
| TWI259657B (en) * | 2004-04-28 | 2006-08-01 | Via Tech Inc | P-domino output latch with accelerated evaluate path |
| CN100550639C (en) * | 2005-10-14 | 2009-10-14 | 威盛电子股份有限公司 | Domino output latch |
| US7990180B2 (en) * | 2009-09-09 | 2011-08-02 | Via Technologies, Inc. | Fast dynamic register |
| KR101911060B1 (en) | 2012-03-19 | 2018-10-23 | 삼성전자주식회사 | Footer-less np domino logic circuit and apparatuses having the same |
| US10104004B2 (en) | 2012-11-08 | 2018-10-16 | Texas Instruments Incorporated | Openflow match and action pipeline structure |
| US8928377B2 (en) | 2013-04-24 | 2015-01-06 | Via Technologies, Inc. | Scannable fast dynamic register |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5121003A (en) * | 1990-10-10 | 1992-06-09 | Hal Computer Systems, Inc. | Zero overhead self-timed iterative logic |
| US5825208A (en) * | 1996-05-20 | 1998-10-20 | International Business Machines Corporation | Method and apparatus for fast evaluation of dynamic CMOS logic circuits |
| US6040716A (en) * | 1997-05-19 | 2000-03-21 | Texas Instruments Incorporated | Domino logic circuits, systems, and methods with precharge control based on completion of evaluation by the subsequent domino logic stage |
| US6239612B1 (en) * | 1997-08-20 | 2001-05-29 | Altera Corporation | Programmable I/O cells with multiple drivers |
| GB2336487B (en) * | 1998-04-17 | 2002-01-09 | Advanced Risc Mach Ltd | Conditional invert functions in precharged circuits |
| JP2016523399A (en) * | 2013-06-28 | 2016-08-08 | インテル コーポレイション | Adaptive interrupt coalescing for energy efficient mobile platforms |
-
2001
- 2001-01-25 US US09/769,075 patent/US6420904B1/en not_active Expired - Fee Related
-
2002
- 2002-01-18 CN CN02800816.2A patent/CN1582532A/en active Pending
- 2002-01-18 JP JP2002560278A patent/JP2004523950A/en active Pending
- 2002-01-18 WO PCT/IB2002/000188 patent/WO2002060061A2/en not_active Application Discontinuation
- 2002-01-18 KR KR1020027012660A patent/KR20030003702A/en not_active Withdrawn
- 2002-01-22 TW TW091100937A patent/TW535355B/en active
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102012111414A1 (en) * | 2012-11-26 | 2014-05-28 | Infineon Technologies Ag | memory circuit |
| US8901979B2 (en) | 2012-11-26 | 2014-12-02 | Infineon Technologies Ag | Storage circuit |
| DE102012111414B4 (en) | 2012-11-26 | 2018-10-04 | Infineon Technologies Ag | Master-slave memory circuit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1582532A (en) | 2005-02-16 |
| WO2002060061A2 (en) | 2002-08-01 |
| JP2004523950A (en) | 2004-08-05 |
| WO2002060061A3 (en) | 2003-12-31 |
| KR20030003702A (en) | 2003-01-10 |
| TW535355B (en) | 2003-06-01 |
| US6420904B1 (en) | 2002-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7696649B2 (en) | Power control circuitry and method | |
| US8890573B2 (en) | Clock gating latch, method of operation thereof and integrated circuit employing the same | |
| US8341578B2 (en) | Clock gater with test features and low setup time | |
| US6087855A (en) | High performance dynamic multiplexers without clocked NFET | |
| US6429689B1 (en) | Method and apparatus for controlling both active and standby power in domino circuits | |
| US5521538A (en) | Adiabatic logic | |
| US6133759A (en) | Decoupled reset dynamic logic circuit | |
| US20030189445A1 (en) | Circuits and systems for limited switch dynamic logic | |
| US20080116938A1 (en) | Hybrid Keeper Circuit for Dynamic Logic | |
| US6046608A (en) | Differential precharge circuit | |
| US6420904B1 (en) | Domino logic with self-timed precharge | |
| US6646487B2 (en) | Method and system for reducing hazards in a flip-flop | |
| US5986475A (en) | Apparatus and method for resetting a dynamic logic circuit | |
| US6914453B2 (en) | Integrated logic and latch design with clock gating at static input signals | |
| US6960939B2 (en) | Limited switch dynamic logic circuit with keeper | |
| US6407585B1 (en) | Method and apparatus for a family of self clocked dynamic circuits | |
| US6373290B1 (en) | Clock-delayed pseudo-NMOS domino logic | |
| US6518796B1 (en) | Dynamic CMOS circuits with individually adjustable noise immunity | |
| US6888377B2 (en) | Duo-mode keeper circuit | |
| US7202705B2 (en) | Dynamic logic circuit apparatus and method for reducing leakage power consumption via separate clock and output stage control | |
| US7193446B2 (en) | Dynamic logic circuit incorporating reduced leakage state-retaining devices | |
| US5977789A (en) | Fast-switching logic gate | |
| US5646566A (en) | Low power clocked set/reset fast dynamic latch | |
| US6081135A (en) | Device and method to reduce power consumption in integrated semiconductor devices | |
| US7129754B2 (en) | Controlled load limited switch dynamic logic circuitry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| AS | Assignment |
Owner name: U.S. PHILIPS CORPORATION, NEW YORK Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:DUTTA, SANTANU;SINGH, DEEPAK;REEL/FRAME:011487/0979 Effective date: 20010122 |
|
| AS | Assignment |
Owner name: KONINKLIJKE PHILIPS ELECTRONICS N.V., NETHERLANDS Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:PHILIPS ELECTRONICS NORTH AMERICA CORPORATION;REEL/FRAME:013296/0171 Effective date: 20020521 |
|
| REMI | Maintenance fee reminder mailed | ||
| LAPS | Lapse for failure to pay maintenance fees | ||
| STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
| FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20060716 |