JP4814561B2 - Semiconductor integrated circuit - Google Patents

Description

  The present invention relates to a semiconductor integrated circuit, and more particularly to speeding up the semiconductor integrated circuit.

  2. Description of the Related Art Conventionally, in semiconductor integrated circuits, particularly flip-flop circuits, as described in Patent Document 1, for example, a dynamic circuit is used for the internal configuration to increase the speed. The dynamic flip-flop circuit described in Patent Document 1 has a function of inputting a plurality of data, selecting any one of them, holding and outputting the data.

  The configuration of the flip-flop circuit with the data selection function is shown in FIG. In FIG. 2A, a data selection circuit 91 is disposed in the previous stage of the holding circuit 90. In the data selection circuit 91, during the low period of the clock CLK, the node N1 is precharged to the power supply potential Vdd by the p-type transistor Tr1, and the node N2 is precharged to the power supply potential Vdd by the p-type transistor Tr50. In the vicinity of the end of this period, one of the selection signals S0 to S2 for selecting any one of the plurality of data D0 to D2 is set to High, and then, when the High period of the clock CLK is entered, the selected data When (for example, D0) is High, the charge of the node N1 is discharged through the n-type transistor Tr2, and the potential of the node N1 becomes the ground potential. Along with this, the n-type transistor Tr51 is turned off, and the precharge potential of the node N2 is held, and this potential is held as the H value by the holding circuit 90 and output as the output signal Q having the H value.

  On the other hand, when the selected data D0 is Low, the charge of the node N1 is not discharged, the potential of the node N1 maintains the precharge potential, and the n-type transistor Tr51 is turned on. As a result, the electric charge at the node N2 is discharged through the n-type transistor Tr51 and the n-type transistor Tr2 to become an L value, and this L value is held by the holding circuit 90 and output as an output signal Q having an L value. .

In FIG. 3A, SI is a data input at the time of scanning, SE is a scan shift control signal, and SEB is an inverted signal of the scan shift control signal.
Japanese Patent Application Laid-Open No. 2003-060497

  However, it has been found that the conventional dynamic flip-flop circuit with a data selection function has a drawback that malfunction occurs when none of a plurality of data is selected. Hereinafter, this malfunction will be described.

  During normal operation, for example, when the node N2 is at the precharge potential (H value) and the output signal Q having the H value is output from the holding circuit 90, a plurality of data is output in the High period of the next cycle of the clock CLK. When all of D0 to D2 are not selected, that is, when all the selection signals S0 to S2 are low values, the n-type transistor Tr2 is turned on, but the precharge potential of the node N1 is held, and the n-type transistor Since the transistor Tr51 is turned on, the charge at the node N2 is discharged through these n-type transistors Tr51 and Tr2 to become an L value, and an output signal Q having an L value is erroneously output from the holding circuit 90. Become.

  In order to eliminate such drawbacks, for example, as shown in FIG. 3B, it is conceivable to add the following circuit as a signal input to the gate of the n-type transistor Tr2. That is, the circuit 92 includes an OR circuit that inputs all the selection signals S0 to S2, a latch circuit that latches the output of the OR circuit in the High period of the clock CLK, the output of the latch circuit, and the clock CLK. It is conceivable that a static circuit composed of an AND circuit 93 for inputting the AND circuit 93 is separately provided and the output of the AND circuit 93 is input to the gate of the n-type transistor Tr2.

  However, in this idea, since all the selection signals S0 to S2 need to pass through the OR circuit and the latch circuit before the rising of the clock CLK, an extra setup time (the static static signal before the rising edge of the clock CLK). Time required to determine the output of the circuit), and there is a disadvantage that hinders the high-speed operation.

  The present invention has been made in view of the above-described problems, and its object is to provide a dynamic flip-flop circuit with a data selection function in a case where none of a plurality of data is selected while ensuring good high-speed operation. Even so, it is to allow normal operation.

  In order to achieve the above object, in the present invention, when none of a plurality of data is selected, for example, in the semiconductor integrated circuit of FIG. To prevent this, the H value of the node N2 is held, and is held and output as it is by the holding circuit.

  That is, in the semiconductor integrated circuit according to the first aspect of the present invention, a clock, a plurality of data, and a plurality of selection signals for selecting the respective data are input, and when the clock transitions, one selected by the selection signal In the semiconductor integrated circuit that outputs data to the holding circuit, the non-selected state detecting circuit includes a non-selected state detecting circuit that detects a state in which all of the plurality of selection signals do not select any of the plurality of data. When a state in which all of the plurality of selection signals do not select any of the plurality of data is detected, the output data of the holding circuit is held by preventing a change in the previously selected data. To do.

  According to a second aspect of the present invention, in the semiconductor integrated circuit, the first clock and a plurality of data are input, and one of the period from the rising edge to the falling edge and the period from the falling edge to the rising edge of the first clock is input. In the first period, which is the period, the first output node is charged, and in the second period, which is the other period, when the values of the plurality of data all match, the charge of the first output node is On the other hand, when at least one of the plurality of data does not match, the NOR-type first dynamic circuit that discharges the charge of the first output node, the second clock, and the first dynamic circuit of the first dynamic circuit When a signal at one output node is input and the charge at the first output node of the first dynamic circuit is discharged in the first period or the second period of the second clock. A semiconductor device comprising: a NAND-type second dynamic circuit that retains the charge of the second output node and discharges the charge of the second output node when the charge of the first output node is retained In the integrated circuit, a third clock and a plurality of selection signals for selecting each of the plurality of data are input, and the charge is charged to the third output node in the first period of the third clock, and the second In a period, when all of the plurality of selection signals do not select any of the plurality of data, a NOR type third dynamic circuit that holds the charge of the third output node, a fourth clock, A signal of the third output node of the third dynamic circuit is input, and the third output node of the third dynamic circuit is input in the first period or the second period of the fourth clock. And a NAND type fourth dynamic circuit that discharges the charge of the fourth output node when the load is held, and the second dynamic circuit receives the signal of the fourth output node of the fourth dynamic circuit as an input When the charge at the fourth output node is discharged, the charge at the second output node is held even when the charge at the first output node of the first dynamic circuit is held. It is characterized by.

  According to a third aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the NOR-type third dynamic circuit and the NAND-type fourth dynamic circuit are the NOR-type first dynamic circuit in terms of physical arrangement. Rather, it is closer to the NAND type second dynamic circuit.

  According to a fourth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the NOR-type third dynamic circuit and the NAND-type fourth dynamic circuit are faster than the NOR-type first dynamic circuit. It is characterized by operation.

  According to a fifth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the NOR-type third dynamic circuit and the NAND-type fourth dynamic circuit have a supply voltage higher than that of the NOR-type first dynamic circuit. Is characterized by high.

  According to a sixth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the NOR-type third dynamic circuit and the NAND-type fourth dynamic circuit are element isolations formed on a semiconductor substrate in physical arrangement. The distance from the region is at a position farther than the NOR type first dynamic circuit.

  According to a seventh aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the semiconductor integrated circuit further includes an output circuit that receives the selected data and outputs the selected data, and the output circuit includes the NOR-type first dynamic circuit. A NOR circuit that receives an output of the circuit and an inverted output of the output of the NAND-type second dynamic circuit; a first n-type transistor that receives the output of the NOR circuit at a gate; and the NAND-type second dynamic circuit A first p-type transistor having a circuit output input to a gate is included, and the first n-type transistor and the first p-type transistor have drains connected to each other.

  According to an eighth aspect of the present invention, in the semiconductor integrated circuit according to the seventh aspect, the output circuit further includes a second n-type transistor in which an output of the NAND-type fourth dynamic circuit is input to a gate. The drain of the second n-type transistor is connected to the source of the first n-type transistor.

  According to a ninth aspect of the present invention, in the semiconductor integrated circuit according to the seventh aspect, the drain of the first n-type transistor and the drain of the first p-type transistor, which are two output terminals of the output circuit, A holding circuit for holding the selection data is connected.

  A tenth aspect of the present invention is the semiconductor integrated circuit according to the ninth aspect, wherein the holding circuit includes a first inverter circuit in which a drain of the first p-type transistor of the output circuit is connected to an input side. The output of the first inverter circuit is input, the second inverter circuit formed by connecting the first n-type transistor and the first P-type transistor in series, and the output of the NAND-type second dynamic circuit. A second n-type transistor that receives the second n-type transistor, and the second n-type transistor is between the n-type transistor and the p-type transistor of the second inverter circuit or the n-type transistor of the second inverter circuit. It is characterized by being arranged between and ground.

  The invention according to claim 11 is the semiconductor integrated circuit according to claim 2, further comprising an output circuit for receiving the selected data and outputting the selected data. The output circuit includes a differential input terminal and a differential input terminal. A differential circuit having an output terminal; an OR circuit to which an output of the NOR-type first dynamic circuit and an inverted output of the output of the NAND-type second dynamic circuit are input; The output is input to one of the differential input terminals of the differential circuit, and the output of the second NAND dynamic circuit is input to the other differential input terminal of the differential circuit. To do.

  According to a twelfth aspect of the present invention, in the semiconductor integrated circuit according to the eleventh aspect, the differential circuit includes a control transistor that enables a differential amplification operation of the differential circuit. The drain is connected to the source of the differential circuit, the source is grounded, and the gate is constituted by an n-type transistor receiving a control signal.

  According to a thirteenth aspect of the present invention, in the semiconductor integrated circuit according to the twelfth aspect, a resistance element is connected in parallel to the control transistor.

  According to a fourteenth aspect of the present invention, in the semiconductor integrated circuit according to the twelfth aspect of the present invention, the semiconductor integrated circuit further includes a signal generation circuit that generates a control signal to be applied to the gate of the control transistor. A short pulse generation circuit to generate, and a NAND circuit to which the short pulse and the output of the NAND type fourth dynamic circuit are input, and the output of the NAND circuit is the gate of the control transistor as the control signal It is characterized by being given to.

  According to a fifteenth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the NOR type third dynamic circuit and the NAND type fourth dynamic circuit have threshold voltages of transistors constituting them. The threshold voltage of the transistors constituting the NOR type first dynamic circuit is lower than the threshold voltage.

According to a sixteenth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the first, second, third and fourth clock signals are shared by one clock signal .

According to a seventeenth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, an inversion node of the third output node of the third dynamic circuit is connected to the second dynamic circuit, and the second dynamic circuit is connected. The circuit is configured to charge the second output of the second dynamic circuit when the inverted node of the third output node is charged and the charge of the fourth output node of the fourth dynamic circuit is held. When the charge at the output node is discharged, the charge at the inversion node of the third output node is held and the charge at the fourth output node is discharged, the charge at the second output node is held. It is characterized by that.

According to an eighteenth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the third dynamic circuit includes a first n-type transistor to which the third clock signal is input to a gate; a second n-type transistor group whose source is commonly connected to the drain of the n-type transistor; and a third n-type transistor group whose source is commonly connected to the drain of the second n-type transistor group. The potential of at least one of the gates of the second n-type transistor group is set to the power supply potential, the potential of the other gate is set to the ground potential, and the gate of the third n-type transistor group is set to Are connected to any one of the plurality of selection signals, and drains of the third n-type transistor group are commonly connected to the third output node, and An inversion node of the third output node of the third dynamic circuit is connected to the path, and the fourth dynamic circuit is shared by the second n-type transistor group in addition to the third output node. An inversion node of the drain connected to is connected, and the second dynamic circuit is charged with an electric charge at an inversion node of the third output node and an electric charge of the fourth output node of the fourth dynamic circuit Is held, the charge of the second output node of the second dynamic circuit is discharged, the charge of the inverted node of the third output node is held, and the charge of the fourth output node is When discharged, the charge of the second output node is held.

According to a nineteenth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the third dynamic circuit has a common source with the first n-type transistor to which the third clock signal is input to the gate. And a gate of the third n-type transistor group is connected to one of the plurality of selection signals, and the third n-type transistor group has a gate connected to the third n-type transistor group. The drain is commonly connected to the third output node, and the fourth dynamic circuit discharges the charge of the fourth output node when the charge of the third output node is held. When the charge of the third output node is discharged, the charge of the fourth output node is held.

According to a twentieth aspect of the present invention, in the semiconductor integrated circuit according to the eighteenth or nineteenth aspect , the third dynamic circuit is configured such that the third clock signal is input to a gate and a drain is the third n-type transistor group. The first p-type transistor connected to the source of the first n-type transistor, the third clock signal is input to the gate, the source is connected to the drain of the third n-type transistor group, and the drain is the third n-type transistor The semiconductor device further includes a second p-type transistor connected to a source of the transistor group.

According to a twenty- first aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the charge charged in the first output node is supplied from the third output node.

According to a twenty- second aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the charge charged in the second output node is supplied from the first output node.

According to a twenty- third aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the charge charged in the fourth output node is supplied from the first output node.

According to a twenty-fourth aspect of the present invention, in the semiconductor integrated circuit according to the second aspect, the first dynamic circuit has a gate connected to an inverted output of the first output node of the first dynamic circuit. a p-type transistor and a second p-type transistor having a gate connected to the first clock signal, wherein the first p-type transistor and the second p-type transistor are connected in series; The source of the type transistor is connected to a power supply, and the drain of the other p-type transistor is connected to the fourth output node or the second output node.

According to a twenty-fifth aspect of the present invention, in the semiconductor integrated circuit of the eighteenth aspect, the potential of at least one of the gates of the second n-type transistor group is connected to a power supply potential via a potential setting transistor. The second n-type transistor group and the potential setting transistor are arranged in the same standard cell.

According to a twenty-sixth aspect of the present invention, in the semiconductor integrated circuit according to the twenty-fifth aspect , the potential setting transistor has a drain connected to at least one gate of the gates of the second n-type transistor group. In the same standard cell, an n-type transistor having a source grounded and a drain and a gate connected to the potential setting transistor is provided in the same standard cell.

The invention according to claim 27 is a semiconductor integrated circuit comprising two semiconductor integrated circuits according to claim 7, wherein the sources and drains of the respective first n-type transistors are connected in common, The sources and drains of the first p-type transistors are connected in common.

A twenty-eighth aspect of the invention is a semiconductor integrated circuit including two semiconductor integrated circuits according to the seventh aspect, wherein logic is formed using each of the output circuits.

According to a twenty- ninth aspect of the present invention, in the semiconductor integrated circuit according to the twenty- eighth aspect, the first inverter circuit in which the drains of the first p-type transistors included in the respective output circuits are connected in common; A holding circuit is configured by the second inverter circuit to which the output of the first inverter circuit is input, and between the n-type transistor and the p-type transistor constituting the second inverter circuit, or the n-type transistor An n-type transistor having a common gate and the corresponding first p-type transistor is arranged in series between each of the transistors and the ground.

The invention of claim 30 is a semiconductor integrated circuit comprising the semiconductor integrated circuit of claim 7 and a static flip-flop, wherein the output circuit receives an output of the static flip-flop, One of the selected data and the output of the static flip-flop is output.

A thirty- first aspect of the invention is the semiconductor integrated circuit according to the thirty -third aspect, wherein scan test data is input to the static flip-flop.

As described above, in the invention according to claims 1 to 31 , in the dynamic flip-flop circuit with a data selection function, when the output signal of the data selection circuit is, for example, High, none of the selection signals is activated after that. Even if all the data is not selected, this state is detected and the output signal of the data selection circuit is held high, so that no malfunction occurs .

As described above, according to the semiconductor integrated circuit of the first to thirty-first aspects of the present invention, in the dynamic flip-flop circuit with a data selection function, all the selection signals are not activated and all data Even when the signal is not selected, the output signal of the data selection circuit can be held at the previous value, and the output signal of the holding circuit can be favorably held at the previous value .

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 shows a semiconductor integrated circuit according to Embodiment 1 of the present invention.

  In the figure, D0, D1, and D2 are data, S0, S1, and S2 are selection signals for selecting the data D0 to D2, respectively, SI is a data input during scanning, and SE is a scan shift operation. A scan shift control signal, SEB is an inverted signal of the scan shift control signal, and Q and SO are output terminals.

  The semiconductor integrated circuit shown in FIG. 1 includes a NOR-type first dynamic circuit 1A, a NAND-type second dynamic circuit 1B, a NOR-type third dynamic circuit 1C, a NAND-type fourth dynamic circuit 1D, An output circuit 1E and a holding circuit 1F are included, and the output circuit 1E and the holding circuit 1F constitute a dynamic flip-flop circuit.

  The NOR-type first dynamic circuit 1A receives the three data D0 to DD2, the three selection signals S0 to S2, and the first clock CLK1, and two p-type MOS transistors Tr1 and Tr3. And an n-type MOS transistor Tr2.

  In the first dynamic circuit 1A, all the selection signals S0 to S2 are controlled to be low in the first period of Low, which is a half cycle from the falling edge to the rising edge of the first clock CLK1. Therefore, in this first period, the p-type transistor Tr1 is turned on, and the first output node N1 is precharged to the power supply voltage Vdd. Thereafter, in the High second period which is a half cycle from the rising edge to the falling edge of the first clock CLK1, the p-type transistors Tr1 and Tr3 are turned off and the n-type transistor Tr2 is turned on. One of the selection signals S0 to S2 is controlled to be High. Therefore, in the second period, the potential of the first output node N1 is determined according to the value of one data D0 to D2 selected by the selection signal controlled to be High. For example, when the data D0 is low, the precharge state of the first output node N1 is maintained, and the first output node N1 is maintained at the power supply potential Vdd, while when the data D0 is high. The electric charge of the first output node N1 is discharged through the n-type transistor Tr2, and the first output node N1 becomes the ground potential.

  The NAND-type second dynamic circuit 1B receives the second clock CLK2 and the signal of the first output node N1 of the first dynamic circuit 1A. Further, the NAND-type second dynamic circuit 1B includes two p-type MOS transistors Tr4 and Tr8 and three n-type MOS transistors Tr5 to Tr7. The signal of the first output node N1 of the first dynamic circuit 1A is input to the gate of the n-type transistor Tr5.

  In the second dynamic circuit 1B, since the p-type transistor Tr4 is on and the n-type transistor Tr7 is off during the first period when the second clock CLK2 is low, the second output node N2 is set to the power supply potential Vdd. Precharged. Thereafter, in the second period in which the second clock CLK2 is High, the precharge operation is stopped and the n-type transistor Tr5 is turned on according to the potential of the first output node N1 of the first dynamic circuit 1A. Turn off. For example, when the Lo data D0 is selected, the n-type transistor Tr5 is turned off and the precharge state of the second output node N2 is maintained, while when the High data D0 is selected. The n-type transistor Tr5 is turned on, and the charge at the second output node N2 is discharged through the two n-type transistors Tr5 and Tr7. The n-type transistor Tr6 is an important transistor in the first embodiment, and its function will be described later.

  Further, the NOR type third dynamic circuit 1C receives the third clock CLK3, the three selection signals S0 to S2 and the scan shift control signal SE, and two p-type transistors Tr9 and Tr11. And an n-type transistor Tr10 and an inverter IN3.

  The third dynamic circuit (non-selection state detection circuit) 1C has a third output because the p-type transistor Tr9 is turned on and the n-type transistor Tr10 is turned off in the first period when the third clock CLK3 is Low. Node N3 is precharged to power supply potential Vdd. Thereafter, in the second period in which the third clock CLK3 is High, when the three selection signals S0 to S2 and the scan shift control signal SE are all Low, that is, in a state where all the data D0 to D2 are not selected. When the precharge state of the third output node N3 is held and this state is detected, when any one selection signal transitions to High, the charge of the third output node N3 is n-type. It is discharged through the transistor Tr10 and its potential becomes Low.

  In addition, the NAND-type fourth dynamic circuit 1D receives the fourth clock CLK4 and the signal of the third output node N3 of the third dynamic circuit 1C, and two p-type MOS transistors Tr12. Tr15 and two n-type MOS transistors Tr13 and Tr14. A signal of the third output node N3 of the third dynamic circuit 1C is input to the gate of the n-type MOS transistor Tr13.

  In the NAND type fourth dynamic circuit 1D, the p-type transistor Tr12 is turned on and the n-type MOS transistor Tr14 is turned off in the first period in which the fourth clock CLK4 is Low. Precharged to power supply potential Vdd. On the other hand, in the second period in which the fourth clock CLK4 is High, the p-type transistor Tr12 is turned off, so that the precharge operation is stopped and the n-type MOS transistor Tr14 is turned on. The potential of the fourth output node N4 is determined according to whether the Tr13 is on or off. That is, in the second period, the charge of the third output node N3 of the third dynamic circuit 1C is held. That is, any data D0 to D2 is kept in a low state during the normal operation. When D2 is not selected, the charge at the fourth output node N4 is discharged through the n-type transistors Tr13 and Tr14, and the potential becomes Low, while one High signal is generated by any one of the selection signals S0 to S2. When the state data is selected and the charge of the third output node N3 of the third dynamic circuit 1C is discharged, the n-type MOS transistor Tr13 is turned off and the precharge state of the fourth output node N4 Is retained.

  The NAND-type second dynamic circuit 1B includes an n-type MOS transistor Tr6 that receives the signal of the fourth output node N4 of the fourth dynamic circuit 1D at its gate. The n-type transistor Tr6 is connected in series with the n-type transistor Tr5. When the n-type transistor Tr6 is in the off state even when the n-type transistor Tr5 is in the on state, the second output node The charge of N2 is not discharged and the precharged state is maintained.

  In the present embodiment, in the NAND-type second dynamic circuit 1B, when all the data D0 to D2 are not selected, the n-type transistor Tr6 is already in the off state when the n-type transistor Tr5 is in the on state. However, as this configuration, the third and fourth dynamic circuits 1C and 1D that control the n-type transistor Tr6 have a configuration that operates faster than the first dynamic circuit 1A. For example, in the third dynamic circuit 1C, the number of series stages of transistors on the path from the third output node N3 to the ground is two, while in the first dynamic circuit 1A, the first output node N1 reaches the ground. Since the number of series stages is 3, the operating speed of the third dynamic circuit 1C is faster than that of the first dynamic circuit 1A. The third and fourth dynamic circuits 1C and 1D are arranged closer to the second dynamic circuit 1B than the first dynamic circuit 1A. Thereby, the delay time in which the potential change of the third and fourth output nodes N3 and N4 of the third and fourth dynamic circuits 1C and 1D propagates to the n-type transistor Tr6 of the second dynamic circuit 1B is shortened, and the first The potential change at the first output node N1 of the first dynamic circuit 1A is shorter than the delay time for propagating to the n-type transistor Tr5 of the second dynamic circuit 1B.

  Further, in order to operate the third and fourth dynamic circuits 1C and 1D at a higher speed than the first dynamic circuit 1A, the supply voltage to the third and fourth dynamic circuits 1C and 1D is set higher than the supply voltage of the first dynamic circuit 1A. The threshold voltage of the MOS transistors constituting the third and fourth dynamic circuits 1C, 1D is set lower than the threshold voltage of the MOS transistors constituting the first dynamic circuit 1A, or the third and fourth The size of the MOS transistors constituting the dynamic circuits 1C and 1D may be set larger than the size of the MOS transistors constituting the first dynamic circuit 1A. Further, when an STI (Shallow Trench Isolation element isolation region) is formed on a semiconductor substrate, there is a concern that the performance of the transistor may deteriorate due to the influence of the STI. Therefore, an arrangement configuration in consideration of the influence of the STI is adopted. You may do it. For example, as shown in FIG. 2, when a transistor array 61 is formed on an N-type substrate 60, a plurality of transistors located at the end of the transistor array 61 are used to form the first dynamic circuit 1 </ b> A. An n-type transistor is configured, and on the other hand, the n-type transistors of the third and fourth dynamic circuits 1C and 1D are configured by using a plurality of transistors located inside the transistor row 61. By adopting this configuration, on the N-type substrate 60, an element isolation region (STI) 65 is arranged between the transistor rows 62 and 63 located on the left and right of the transistor row 61, and is affected by this STI. The transistor at the end of the transistor array 61 has a large degree of deterioration, but since the n-type transistor of the first dynamic circuit 1A that does not need to be operated at high speed is arranged, the influence of the deterioration is small. On the other hand, the n-type transistors of the third and fourth dynamic circuits 1C and 1D, which require a high operating speed, are composed of transistors that are located inside the transistor array 61 and are not easily affected by STI. Good operating speed can be secured.

  In the present embodiment, the operation speed of the third and fourth dynamic circuits 1C and 1D is configured to be faster than that of the first dynamic circuit 1A. However, the present invention includes a case where this configuration is not adopted. For example, in the semiconductor integrated circuit of FIG. 1, the second clock CLK2 is input to the gate of the n-type transistor Tr7 of the second dynamic circuit 1B. Instead of the second clock CLK2, the third clock of the third dynamic circuit 1C is used. An inverted signal of the signal at the third output node N3 may be input. Under this configuration, when all data is not selected (all selection signals S0 to S2 are Low) before the rise of the fourth clock CLK4, the third output node N3 becomes High, When the n-type transistor Tr7 is turned off and thereafter the fourth clock CLK4 rises and changes, the fourth output node N4 becomes low and the n-type transistor Tr6 is turned off. Therefore, the third and fourth dynamic circuits 1C and 1D It is not necessary to configure the operation speed faster than that of the first dynamic circuit 1A.

  Next, the output circuit 1E and the holding circuit 1F constituting the remaining part of the dynamic flip-flop circuit will be described. The output circuit 1E receives the signal of the first output node N1 of the first dynamic circuit 1A and the signal of the second output node N2 of the second dynamic circuit 1B, and also includes an inverter IN4 and a NOR circuit. NOR1, a first p-type MOS transistor Tr20, and three n-type MOS transistors Tr21, Tr22, Tr23, and the drain of the p-type MOS transistor Tr20 and the drain of the first n-type transistor Tr21 are mutually connected. The seventh output node N7 is connected. A signal of the second output node N2 of the second dynamic circuit 1B is input to the gate of the p-type MOS transistor Tr20. The NOR circuit NOR1 includes two p-type transistors Tr24 and Tr25 and an n-type transistor Tr26, and the signal of the first output node N1 of the first dynamic circuit 1A and the second dynamic circuit 1B. A signal obtained by inverting the signal of the second output node N2 by the inverter IN4 is input, and the output is input to the gate of the first n-type transistor Tr21 as the sixth output node N6.

  Therefore, in the output circuit 1E, when the second output node N2 of the second dynamic circuit 1B is in the low state and the first output node N1 of the first dynamic circuit 1A is in the high state, the p-type transistor Tr20 is turned on. Since the n-type transistor Tr21 is turned off, the seventh output node N7 is precharged to the power supply potential Vdd, and the potential becomes High. On the other hand, when the second output node N2 is in the high state and the first output node N1 is in the low state, the p-type transistor Tr20 is turned off and the n-type transistor Tr21 is turned on, so that the charge of the seventh output node N7 is charged. Is discharged and its potential becomes Low.

  In the output circuit 1E, the signal of the fourth output node N4 of the fourth dynamic circuit 1D is input to the gate of the second n-type transistor Tr22, the source is grounded, and the drain is the n-type transistor. Connected to the source of Tr21. When the potential of the seventh output node N7 is High, the n-type transistor Tr22 outputs (the sixth output) of the NOR circuit NOR1 due to the potential drop of the first output node N1 of the first dynamic circuit 1A. Even if the output node N6) is High and the n-type transistor Tr21 is turned on, the potential of the seventh output node N7 malfunctions to Low or a through current flows by maintaining the OFF state of the n-type transistor Tr22. This is to prevent this.

  Next, the holding circuit 1F will be described. The holding circuit 1F functions as a feedback buffer. The first output IN5 is connected to the input side of the seventh output node N7 of the holding circuit 1E, and the inverter IN5 is connected to the input side. And a second inverter IN6. The output side of the second inverter IN6 is connected to the seventh output node N7. Further, the holding circuit 1F includes a second n-type MOS transistor Tr29 arranged in series between the first p-type MOS transistor Tr27 and the first n-type MOS transistor Tr28 constituting the second inverter IN6. And a delay cell 59. Both the inverters IN5 and IN6 hold the potential of the seventh output node N7 of the holding circuit 1E, and this holding potential is inverted by the inverter IN7 and then externally output from the output terminal Q. The output of the first inverter IN5 is output from the output terminal SO after being delayed by the delay cell 59 for a set time.

  In the holding circuit 1F, the n-type MOS transistor Tr29 has the gate thereof supplied with the signal of the second output node N2 of the second dynamic circuit 1B, the drain connected to the drain of the p-type transistor Tr27, and the source thereof. Is connected to the drain of the n-type transistor Tr28. This n-type transistor Tr29 has the following functions. That is, when the seventh output node N7 of the output circuit 1E is Low, the output node N7 is grounded through the n-type transistor Tr28 of the second inverter IN6. When the second output node N2 starts to transition from High to Low, the p-type transistor Tr20 of the output circuit 1E is turned on to start precharging the seventh output node N7 to the power supply potential Vdd. At this time, the n-type transistor Tr29 is turned off by the low state of the second output node N2, and a path grounded from the seventh output node N7 through the n-type transistor Tr28 is cut off, and the seventh output node N7. The precharge operation is promoted.

  Next, the operation of the semiconductor integrated circuit of this embodiment will be described based on the timing chart of FIG. In order to simplify the description, a case where each of the first to fourth clocks CLK1 to CLK4 is the same clock CLK will be described.

  First, in the first period of the clock CLK, the data D0 is High during the data valid period (time required to satisfy the setup and hold times) before and after the clock rise, and goes Low after the data valid period. The other data D1 and D2 are High, and the selection signal S0 is Low during the data valid period, and becomes High after the data valid period has passed. The other selection signals S1 and S2 are Low. Therefore, in this first cycle, no data D0 to D2 is selected.

  In this state, since the first output node N1 of the first dynamic circuit 1A is High during the data valid period, the n-type transistor Tr5 is turned on in the second dynamic circuit 1B. As a result, in the conventional example shown in FIG. 3A, when the second output node N2 is in the High state, the second output node N2 malfunctions to Low, and the original “H” signal is output from the flip-flop circuit to “L”. "The signal was output incorrectly.

  However, in the present embodiment, the third output node N3 of the third dynamic circuit 1C becomes High, and the fourth output node N4 of the fourth dynamic circuit 1D becomes Low after the rising of the clock, so that the second dynamic circuit 1B Then, the n-type transistor Tr6 is turned off at the stage before the n-type transistor Tr5 is turned on, and the second output node N2 is prevented from malfunctioning to Low, and is held at High. Accordingly, in the output circuit 1E, the seventh output node N7 is held low, and a correct “H” signal is output from the holding circuit 1F.

  On the other hand, when the seventh output node N7 of the output circuit 1E is held high, after the rising of the clock CLK, for example, the selection signal S2 temporarily becomes high, and the first output node of the first dynamic circuit 1A Even when N1 becomes Low (not shown), in the output circuit 1E, the sixth output node N6 becomes High and the n-type transistor Tr21 is turned on, but the n-type transistor Tr22 is turned off. The output node N7 is not grounded, and the high level of the seventh output node N7 is maintained. Note that the n-type transistor Tr22 is turned off even if the third output node N3 of the third dynamic circuit 1C becomes Low with the transition of the selection signal S2 to High. This is because the output node N4 is held low.

  In FIG. 4, when the data D0 is Low, the data D1 and D2 are High, the selection signal S0 is High, and the other selection signals S1 and S2 are Low in the second cycle of the clock CLK, that is, the data D0 is selected. It shows the state being done.

  Here, in the present embodiment, as shown in FIG. 3B, since an OR circuit or a latch circuit is not added to the previous stage of the clock, setup of a selection signal is unnecessary, and a dynamic flip-flop that operates at high speed Circuit can be provided.

  In the above description of the operation, the first to fourth clocks CLK1 to CLK4 have been described as being the same clock at the same time, but there may be some differences in the phase between the clocks. In that case, the first clock CLK1 is preferably earlier than the second clock CLK2. The third and fourth clocks CLK3 and CLK4 are preferably earlier than the first and second clocks CLK1 and CLK2.

  The clock CLK2 may be generated based on the clock CLK3 of the third dynamic circuit 1C without setting the delay value of the clock CLK2 input to the second dynamic circuit 1B to a predetermined value. A circuit configuration in this case is shown in FIG. In the figure, a separate dynamic circuit A1 is provided. This dynamic circuit A1 has the same number of n-type MOS transistor series circuits as the number of n-type MOS transistors in the first dynamic circuit 1A shown in FIG. The configuration of the parallel circuit section in which a plurality of series circuits are connected in parallel is the same as that of the first dynamic circuit 1A except for the input configuration of the scan signal SE. The output A1-1 of the separately provided dynamic circuit A1 is inverted by the inverter IN10 and input to the n-type transistor Tr7 of the second dynamic circuit 1B.

  The dynamic circuit A1 separately provided in FIG. 9 further includes a clock generation circuit A2 that generates the clock CLK4 input to the fourth dynamic circuit 1D from the clock CLK3 input to the third dynamic circuit 1C of FIG. It is. This clock generation circuit A2 is configured so that the junction capacitance portion of the multi-stage input gate group for data or the like looks equivalent to the point of the output A1-1 of the dynamic circuit A1, and the output A2-1 is inverted by the inverter IN11. And input to the n-type transistor Tr14 of the fourth dynamic circuit 1D. The clock generation circuit A2 is further provided with a precharge circuit A2-2 composed of a p-type MOS transistor Tr40. The precharge circuit A2-2 has a function of precharging the output point A2-1 of the clock generation circuit A2, and the clock input to the gate of the p-type transistor Tr40 is supplied to the third dynamic circuit 1C. This is the input clock CLK3. The delay difference during discharge between the output A1-1 of the dynamic circuit A1 and the output A2-1 of the clock generation circuit A2 reflects the current difference of the n-type MOS transistors to which the selection signals S0 to S3 are input. By compensating for the delay difference with the inverter IN11, a reliable operation can be realized.

  In the circuit of FIG. 1, when any one of the other selection signals S0 to S3 is output in addition to the selection signal SE, the dynamic circuit A1 transitions faster than the dynamic circuit 1A. Then, the output may become undefined. However, in FIG. 9, in the dynamic circuit A1, the five NMOS transistors Ts1 to Ts4 connected in series with the transistors to which the selection signals SE and S0 to S3 are input have their gates grounded and become non-conductive. It has become. Therefore, since the current path discharged from the node A2-1 to the ground is one path through the NMOS transistor Ts5 whose gate is fixed to the power supply potential Vdd, the dynamic circuit A1 transitions later than the dynamic circuit 1A. As a result, the data output to the output terminal Q becomes the OR output of the data selected from the data D0 to D3. This configuration is effective because, for example, the expected value does not become unstable during the scan test.

  FIG. 10 shows a layout configuration example of the semiconductor integrated circuit shown in FIG. In the figure, a circuit portion of an n-type transistor for inputting selection signals S0 to S3 and an n-type transistor for inputting data D0 to D3 of the first dynamic circuit 1A and the dynamic circuit A1 shown in FIG. 9 are configured. The circuit parts of the n-type transistors for inputting the selection signals S0 to S3 are arranged in a vertical arrangement relationship. As a result, the wiring capacity of the input pin is reduced, and since both of these circuit portions are close to each other, the variation component of the manufacturing process between the two dynamic circuits 1A and A1 is also reduced, and the voltage fluctuation And there is an advantage that the temperature fluctuation is reduced. In addition, in the layout of the selection signal and data input circuit unit composed of a plurality of n-type transistors, when creating layouts with different numbers of selection signal and data inputs, the one with the largest number of inputs should be created. For example, in order to create a layout having a smaller number of inputs, it is only necessary to reduce the n-type MOS transistor on the left side of FIG.

  Note that the transistor Tr91 of the dynamic circuit 1A functions as a keeper and holds the charge of the node N1. Here, the source of the transistor Tr91 is preferably connected to the drain (node N20) of the transistor Tr93 of the dynamic circuit A1. This is because, for example, the signal transition speed of the node N1 is faster than when the source of the transistor Tr91 is connected to the drain of the transistor Tr94 of the dynamic circuit 1A. This is because the drain capacitance of the transistor Tr93 of the dynamic circuit A1 does not affect the node N1. The same applies to the transistor Tr92 of the dynamic circuit 1B.

  In addition, when the number of data to be input is very large, it can be considered that these data are divided into two sets. For example, in the semiconductor integrated circuit shown in FIG. 11, the set of the first to fourth dynamic circuits 1A to 1D and A1 shown in FIG. 9 and the other first to fourth dynamic circuits 1A ′ having the same configuration as these. To 1D ′ and A1 ′. When the number of data is 2N, data D1 to SN is input to one set, and data SDN + 1 to D2N is input to the other set. Both of these sets are input in parallel to the gates of the n-type transistors Tr20 and Tr21 of the output circuit 1E shown in FIG. Further, a selection signal coincidence detection circuit 1J for detecting coincidence between the outputs A1-1 and A1-1 ′ of the dynamic circuit A1 and between the outputs A2-1 and A2-1 ′ of the clock generation circuit A2 is further added. 1J-1 is connected to the gate of the n-type transistor Tr22 of the output circuit 1E shown in FIG. With this configuration, the capacity of the first nodes N1 and N1 ′ of the first dynamic circuits 1A and 1A ′ of each set is half that of the case of only one set, so that the operation speed can be increased. is there.

(Embodiment 2)
Next, Embodiment 2 of the present invention will be described. In this embodiment, as shown in FIG. 5, the output circuit 1E shown in FIG. 1 is modified.

  That is, the output circuit 1G shown in FIG. More specifically, the differential circuit 70 has a pair of differential input terminals 70a and 70b and a pair of differential output terminals 70c and 70d, and two cross-coupled p-type MOS transistors. Tr30, Tr31 and two n-type MOS transistors Tr32, Tr33, and two n-type MOS transistors Tr34, Tr35 for differential signal input having the gate connected to the pair of differential input terminals 70a, 70b, Have The pair of differential output terminals 70c and 70d are connected to the connection point between one p-type transistor Tr30 and the n-type transistor Tr32 and the connection point between the other p-type transistor Tr31 and the n-type transistor Tr33. The pair of differential output terminals 70c and 70d become the output terminal Q of the semiconductor integrated circuit in FIG. 1 and its inverted output terminal NQ.

  A signal of the second output node N2 of the second dynamic circuit 1B shown in FIG. 1 is input to the one differential input terminal 70a. On the other hand, an OR circuit 71 is connected to the other differential input terminal 70b. The OR circuit 71 includes a signal obtained by inverting the signal of the second output node N2 of the second dynamic circuit 1B by the inverter 72 and the signal of the first output node N1 of the first dynamic circuit 1A. Entered.

  Furthermore, a control transistor Tr36 composed of an n-type MOS transistor is connected to the ninth node N9 which is the source of the two n-type MOS transistors Tr34 and Tr35 for differential signal input. This control transistor Tr36 has its source grounded, its drain connected to the ninth node N9, and its gate having a fifth clock generated by the clock generation circuit 1H shown in FIG. 6 as a control signal. Signal CLK5 is input.

  The internal configuration of the clock generation circuit 1H will be described. In FIG. 6, a clock generation circuit (signal generation circuit) 1H includes a short pulse generation circuit 75 that generates a short pulse signal in the same cycle as the first clock CLK1, and a NAND circuit NAND11. The short pulse generation circuit 75 includes an inverter IN10 that inverts the first clock CLK1, a NAND circuit NAND10 that receives the outputs of the first clock CLK1 and the inverter IN10, and an inverter IN11 that inverts the output of the NAND circuit. Have The NAND circuit NAND11 receives the output of the inverter IN11 and the signal of the fourth output node N4 of the fourth dynamic circuit 1D shown in FIG. 1, and the output of the NAND circuit NAND11 is the fifth. This clock CLK5 is input as a control signal to one of the differential signal input n-type transistors Tr36 of the differential circuit 70 shown in FIG.

  As shown in FIG. 7, the clock generation circuit 1H shown in FIG. 6 has a period when, for example, the selection signal S1 becomes High and the data D1 is selected in the first cycle of the first clock CLK1. Since the signal of the fourth output node N4 of the fourth dynamic circuit 1D is High at the beginning, when the short pulse is generated from the short pulse generation circuit 75 thereafter, the fifth clock CLK5 is High at this time. Then, the fifth clock CLK5 transitions from Low to High with the end of the short pulse. At this time, the control transistor Tr36 amplifies and outputs the differential input signal by turning on in the middle of the transition of the fifth clock CLK5 from Low to High. In other states, the control transistor Tr36 remains off. Accordingly, when the fifth clock CLK5 is High, the output circuit 70 functions as a latch that holds output data. With this configuration, when the output circuit 1G shown in FIG. 5 is provided, the holding circuit 1F shown in FIG. 1 is not required in the subsequent stage.

  In FIG. 5, in the output circuit 1G, an n-type MOS transistor Tr37 is arranged in parallel with the control transistor Tr36. The n-type transistor (resistance element) Tr37 has a source grounded and a gate and a drain connected to the ninth node N9 of the differential circuit 70. Although the potential of the ninth node N9 may increase due to a leakage current when the fifth clock CLK5 is in a low state, the n-type transistor Tr37 functions as a resistance element to suppress and prevent the potential increase. Therefore, the ninth node N9 is held at an optimum potential state. As a result, the potential between the source and the drain of the pair of n-type transistors Tr34 and Tr35 for differential input is held at an optimum potential difference for obtaining an appropriate gain, so that the desired high-speed operation of the output circuit 1G is good. Maintained.

  In the present embodiment, since the differential circuit 70 quickly amplifies and outputs this in a situation where the potential difference between the input differential signals is small, it is faster than the data holding by the output circuit 1E of the first embodiment.

(Embodiment 3)
FIG. 8 shows a semiconductor integrated circuit according to the third embodiment of the present invention.

  The present embodiment differs from the semiconductor integrated circuit shown in FIG. 1 in the configuration of the NOR-type first dynamic circuit 2A and the NOR-type third dynamic circuit 2C, and the second and fourth dynamic circuits 1B, 1D, The output circuit 1E and the holding circuit 1F are the same as those in the first embodiment.

  In the first dynamic circuit 2A, a p-type transistor Tr1 and an n-type transistor Tr2 are connected in series. The series circuit includes an n-type MOS transistor Tr80 that receives data D at the gate, and an output signal Q at the gate. Is connected in series with another n-type MOS transistor Tr81 receiving the inverted signal NQ. Therefore, in the first dynamic circuit 2A, the potential of the first output node N1 is basically determined according to the value of the data D, and when the data D is output from the output terminal Q, the inverted output NQ Prepared for the next change in the value of data D.

  The third dynamic circuit (coincidence detection circuit) 2C includes an EXNOR circuit EXNOR1. The EXNOR circuit outputs the third output only when the data D and the output signal Q and their inverted signals ND and NQ are input and the data D and the output signal Q coincide after the rising of the third clock CLK3. Node N3 is set to power supply potential Vdd. Therefore, when the data D and the output signal Q coincide with each other, in the fourth dynamic circuit 2D, the n-type transistor Tr13 is turned on, and the charge of the fourth output node N4 is discharged. In the dynamic circuit 2B, the n-type transistor Tr6 is turned off.

  With the above configuration, in the dynamic NAND circuit 2D, when the value of the data D and the value of the output signal Q are the same, the output node N4 transitions to Low, and the n-type transistor Tr6 of the second dynamic circuit 2B is changed. Since it is forcibly turned off, the subsequent operations of the second dynamic circuit 2B, the output circuit 1E, and the holding circuit 1F can be stopped. Therefore, useless operations of these circuits 2B, 1E, and 1F can be prevented, and the power of the semiconductor integrated circuit can be reduced.

  As shown in the first embodiment, the physical arrangement of each dynamic circuit, the size and threshold voltage characteristics of each transistor, the supply voltage to these circuits, and the like are the same as in the first embodiment. Needless to say, it is possible to adopt a configuration. As the output circuit 1E, the differential output circuit 1G as shown in the second embodiment can be adopted. In this case, of course, higher speed can be realized.

  In this embodiment, an example of a flip-flop is shown. However, for example, a latch circuit can be formed by using the potential of the node N2 as an output signal. In this case, the holding circuit 1F does not need to output a signal and need not be provided.

(Embodiment 4)
FIG. 12 is a circuit diagram of another multi-input flip-flop according to the present invention. 1 and FIG. 9 is different in that it operates with a single clock signal CLK1. Furthermore, the difference from the circuits shown in FIGS. 1 and 9 is that a p-type MOS transistor 12B and a p-type MOS transistor 12C are provided.

  In FIG. 1 and FIG. 9, the p-type MOS transistors (transistors Tr4 and Tr12 in FIG. 1) are connected to the power source and charge the nodes N2 and N4. However, in the circuit of FIG. P-type MOS transistors 12B and 12C having their sources and drains connected are inserted between N2 and between nodes N1 and N4, respectively. The gate of the p-type MOS transistor 12B is connected to the node A1_2, and the gate of the p-type MOS transistor 12C is connected to the node A2_3. Since this circuit uses only one clock signal, power consumption can be reduced, and even if only one clock signal is used, no malfunction occurs.

  13 and 14 show the terminals SI, D [1] to D [N-1] and the terminal D [N] or the terminals SE, S [1] to S [N-1] in the circuit shown in FIG. ] Shows the relationship between the voltage of each node and time when the signal input patterns of the terminal S [N] are different. Further, FIGS. 13 and 14 also show waveforms in which a malfunction occurs due to the case where the transistor balance is poor in the circuits of FIGS. 1 and 9 and driving with a single clock signal. In the case of using the circuit, the solid line indicates the case where the circuits of FIGS. 1 and 9 are used.

  The description will be made in comparison with FIG. In FIG. 13, the signal inputs of terminals D [1] to D [N−1], SI, S [1] to S [N], and SE are all set at the timing when the clock signal CLK1 transitions to Hi. The hold time is satisfied and is Lo. Only the terminal DN satisfies the desired setup and hold time and is Hi. Thereafter, only the terminal S [N] transits from Lo to Hi while the clock signal CLK1 is Hi. Then, the node A1-1 and the node N1 transition to Lo, and the node N6 becomes Hi. When the p-type MOS transistor 12C is configured in the same manner as in FIGS. 1 and 9, when the clock signal CLK1 subsequently transitions from Hi to Lo, the power supply voltage Vdd is supplied to the node N4 via the p-type MOS transistor 12C. N4 becomes Hi. As a result, the Hi periods between the node N4 and the node N6 may overlap. When the Hi periods of the node N4 and the node N6 overlap, both the transistors Tr21 and Tr22 are turned on, the charge is extracted from the node N7, and the node N7 originally has to keep Hi. May cause the output terminal Q to malfunction. This is because, in particular, the circuit for controlling the charging of the node N4 and the charging of the node N1 is not devised. Therefore, depending on the variation of the p-type MOS transistor elements charging the nodes N4 and N1, the node N4 is more This is because the battery is charged quickly and causes a malfunction.

  However, in the circuit of FIG. 12, the current characteristic between the drain and the source of the p-type MOS transistor 12C indicates that the voltage between the drain and the source of the p-type MOS transistor 12C is close to the threshold Vtp. A linear characteristic is shown with respect to the voltage between. Further, when the substrate voltage and the source voltage of the p-type MOS transistor 12C are compared, the substrate voltage is higher, so that it behaves as if it is a very high resistance element. That is, since the node N1 is charged first and then the node N4 is easily charged, the timing at which the node N4 becomes Hi is delayed, and the possibility that the nodes N4 and N6 become Hi at the same time is reduced.

  Further, description will be made in comparison with FIG. In FIG. 14, at the timing when the clock signal CLK1 transitions to Hi, the terminal S [N] satisfies the desired setup and hold time and is Hi, and the terminals S [1] to S [N−1], SE, D [1] to D [N]), and the SI signal input satisfies the desired setup and hold time and is Lo. Thereafter, only the terminal D [N] transits from Lo to Hi while the clock signal CLK1 is Hi. Then, the node N1 transitions from Hi to Lo. When the p-type MOS transistor 12B is configured in the same manner as in FIGS. 1 and 9, the nodes N1 and N2 are charged when the clock signal CLK1 subsequently transitions from Hi to Lo, but the node N1 is more than the node N2. If the battery is charged later, the node N2 becomes Hi and the node N1 becomes Lo, so that the node N6 becomes Hi and a Glitch occurs at the node N7. If it is propagated to the output terminal Q, a malfunction may occur.

  However, in the circuit of FIG. 12, due to the p-type MOS transistor 12B, the current characteristic between the drain and source of the p-type MOS transistor 12B is such that the voltage between the drain and source of the p-type MOS transistor 12B is near the threshold value Vtp. Until now, it shows a linear characteristic with respect to the voltage between the drain and the source. When the substrate voltage and the source voltage of the p-type MOS transistor 12B are compared, the substrate voltage is higher, so that it behaves as if it is a very high resistance element. That is, since the node N2 does not become Hi unless the node N1 becomes Hi, the node N6 does not become Hi, and the malfunction is prevented.

  As described above, the source and drain of the p-type MOS transistor 12B are connected to the nodes N1 and N2, respectively, and the source and drain of the p-type MOS transistor 12C are connected to the nodes N1 and N4, respectively. The order of charging between nodes N1 and N4 and the order of charging between nodes N1 and N4 are uniquely determined without being influenced by manufacturing variations in the device size of the p-type MOS transistor, so that a more robust circuit configuration can be obtained. realizable.

  Further, FIG. 12 shows that in the dynamic circuit A1, the gates of the MOS transistors AN, A3 to AN-1 (MOS transistors directly connected to the power supply and ground in FIG. 9) are connected to the two outputs of the circuit 12A. The point is characteristic.

  The circuit 12A includes an n-type MOS transistor 12A-1, a p-type MOS transistor 12A-2, and another n-type MOS transistor 12A-3. The drain of the p-type MOS transistor (potential setting transistor) 12A-2 is connected to the gate of one n-type transistor AN in the second n-type transistor group A3 to AN. Set the gate potential to the power supply potential. The source of the n-type MOS transistor 12A-3 is grounded, and the gate and drain thereof are connected to the gate of the potential setting transistor 12A-2.

  In the miniaturization process, the gate oxide film thickness becomes thin, and the ESD resistance of the gate becomes weak. Therefore, in the circuit as shown in FIG. 9, when an overvoltage is applied to the power supply or the ground, the impedance is low, so that the gate electrode is likely to cause punch-through, and the MOS transistor is easily destroyed. However, as shown in FIG. 12, by providing the circuit 12A, the gate is connected via the resistance between the source and drain of the MOS transistor, so that the impedance from the power supply and the ground becomes high, and the MOS transistor is destroyed. It becomes difficult.

  The circuit 12A may be arranged as a part of a multi-input flip-flop in the same standard cell as the second n-type transistor group A3 to An that operates by inputting the output of the circuit 12A to the gate. desirable. This is because such a multi-input flip-flop has a large number of input terminals, and wiring routing between standard cells becomes complicated. If the circuit 12A does not exist in the cell, a cell such as the circuit 12A needs to be placed somewhere and connected to the multi-input flip-flop by wiring, and the wiring congestion between standard cells becomes more complicated. Because. Furthermore, the wiring between the standard cells is usually performed by automatic placement and routing. Therefore, there is a possibility that the output of the circuit 12A is unintentionally wired so as to be affected by the crosstalk. If the output of the circuit 12A is subjected to crosstalk noise, the flip-flop circuit having a multi-input selection function may cause a malfunction, and therefore, it may be arranged in the standard cell in consideration of the influence of crosstalk as much as possible. desirable.

  In this circuit 12A, the node connected to the gate of the N-type MOS transistor 12A-1 is the drain of the P-type MOS transistor 12A-2 to reduce the elements, but the MOS transistors 12A-2 and 12A-3 In the same manner as in the above configuration, another p-type MOS transistor is prepared, the drain and gate of the p-type MOS transistor are connected in common, and the signal line and the gate of the n-type MOS transistor 12A-1 are connected. May be.

  Further, if the circuit 12A is arranged at the lower ends of the left and right n-type MOS transistors at the bottom of FIG. 10, the circuit A1 and the node N1 in FIG. 12 can be connected to the subsequent stage without lengthening them. If the circuit is a standard cell, NWELL and PWELL are again arranged at the lower end, so that it is possible to arrange them without worrying about the distance constraint of different well boundaries on the lower adjacent boundary between cells.

(Embodiment 5)
FIG. 15 shows a circuit diagram of another multi-input flip-flop of the present invention.

  1 and 9 are different in that they operate with a single clock signal CLK1. Further, the circuits described in FIGS. 1 and 9 are different in a circuit unit 13B in the first dynamic circuit 1E, a circuit unit 13C in the dynamic circuit 1D, and a circuit unit 13A in the dynamic circuit 1A. In FIG. 1, it is a single p-type MOS transistor element (the transistors Tr4 and Tr12 in FIG. 1) whose source is connected to the power source and which charges the dynamic node portions of the nodes N2 and N4. Then, another p-type MOS transistor (p-type MOS transistor 13B1, p-type MOS transistor 13C1) is inserted into the drain of the p-type MOS transistor for further charging, and the nodes N2 and N4 are respectively connected via the source and drain. Connected. The gate of the p-type MOS transistor 13B1 and the gate of the p-type MOS transistor 13C1 are connected to the output of the inverting circuit INV13 at the node N1. Further, the source of the p-type MOS transistor 13A is connected to the power supply in FIG. 11, but is connected to the node A1-1 in FIG. As a result, this circuit uses only one clock signal, so that power consumption can be reduced, and there is a feature that no malfunction occurs even if one clock signal is not used.

  In FIG. 15, an n-type transistor (first n-type transistor) Tr40 is provided in the third dynamic circuit A1. The n-type transistor Tr40 has a gate to which a clock signal CLK1 is input, and a drain connected to the sources of a plurality of n-type transistors (second n-type transistor group) A3 to AN. Further, common sources of a plurality of n-type transistors (third n-type transistor groups) A20 to AK are connected to a common drain of the second n-type transistor groups A3 to AN. A predetermined power supply is connected to the gate of one n-type transistor AN in the second n-type transistor group A3 to AN, and the gate potential is set to the power supply potential. The gates of the other n-type transistors A3 to A5 are all grounded, and the gate potential is set to the ground potential. The plurality of selection signals S [1] to S [N] are input to the gates of the third n-type transistor groups A20 to AK, respectively, and the drains of the third n-type transistor groups A20 to AK are Commonly connected to the third output node N3.

  An inversion node A1-2 of the third output node N3 (= A1-1) of the third dynamic circuit A1 is connected to the second dynamic circuit 1E, and the third dynamic circuit 1D is connected to the third dynamic node 1E. In addition to the output node N3, the inversion node A2-2 of the node A2-1 of the common drain of the second n-type transistor groups A3 to AN is connected.

  FIGS. 16 and 17 illustrate the circuit between the terminals D [1] to D [N−1] and the terminal D [N] of the circuit described in FIG. 15 and the terminals S [1] to S [N−1]. The relationship between the voltage of each node and the time when the signal input pattern is different between the S and the terminal S [N] is shown. Further, FIGS. 16 and 17 also show waveforms that cause a malfunction when the circuit of FIG. 9 is driven by a single clock signal in the case of poor transistor balance, and the alternate long and short dash line indicates the circuit of FIG. When used, the solid line indicates the case where the circuit of FIG. 9 is used.

  The description will be made in comparison with FIG. In FIG. 16, all the input signals of the terminals S [1] to S [N] satisfy the desired setup and hold time at the timing when the clock signal CLK1 transitions to Hi, and are Lo. Thereafter, only the terminal S [N] transits from Lo to Hi while the clock signal CLK1 is Hi. Then, the node A1-1 and the node N1 transition to Lo, and the node N6 becomes Hi. When the circuit 13C is configured in the same manner as in FIGS. 1 and 9, when the clock signal CLK1 subsequently transitions from Hi to Lo, the power supply voltage Vdd is supplied to the node N4 via the p-type MOS transistors 13C1 and 13C2, and the node N4 Becomes Hi. As a result, the Hi periods between the node N4 and the node N6 may overlap. When the Hi period between the node N4 and the node N6 overlaps, both the transistors Tr21 and Tr22 are turned on, the charge is extracted from the node N7, and the node N7 originally has to keep Hi. There is a possibility that the output terminal Q malfunctions. This is because the circuit for controlling the charging of the node N4 and the charging of the node N1 is not particularly devised, so that the node N4 is faster depending on the variation of the p-type MOS transistor elements that charge the nodes N4 and N1, respectively. It is caused by being charged and causing malfunction.

  However, in the circuit of FIG. 15, the circuit 13C is not turned on unless the potential of the output of the inverting circuit INV13 at the node N1 is equal to or less than the difference obtained by subtracting the threshold voltage of the p-type MOS transistor 13C1 in the circuit 13C from the power supply voltage VDD. The node N1 is charged first, and the node N4 is easily charged later. Therefore, the possibility that the nodes N4 and N6 become Hi at the same time is low.

  Further, description will be made in comparison with FIG. In FIG. 17, at the timing when the clock signal CLK1 transitions to Hi, the terminal S [N] satisfies the desired setup and hold time and is Hi, and the terminals S [1] to S [N−1], SE , D [1] to D [N]), the SI input signal is Lo, satisfying the desired setup and hold time. Thereafter, only the terminal D [N] transits from Lo to Hi while the clock signal CLK1 is in the Hi period. Then, the node N1 transitions from Hi to Lo. Thereafter, when the clock signal CLK1 transitions from Hi to Lo, in the circuit of FIG. 1, the node N1 and the node N2 are charged, but when the order of the node N1 is charged after the node N2, Node N2 is Hi and Node N1 is Lo. Therefore, the node N6 becomes Hi, and Glitch is generated at the node N7. If it is propagated to the output terminal Q, a malfunction occurs.

  However, in the circuit of FIG. 15, if the potential of the output of the inverting circuit INV13 at the node N1 is not less than or equal to the difference obtained by subtracting the threshold voltage of the first p-type MOS transistor 13B1 in the circuit 13B from the power supply voltage VDD, the node N2 Since the battery is not charged, the node N2 does not become Hi unless the node N1 becomes Hi. That is, since the node N6 does not become Hi, malfunction is prevented.

  Further, in FIG. 18, when the clock signal CLK1 transits to Hi, the terminals D [N] and S [N] satisfy the desired setup and hold time and are Hi, and the terminals S [1] to S [N −1], SE, D [1] to D [N−1], and SI input signals satisfy the desired setup and hold times and are Lo. Thereafter, the terminal D [N] transits from Hi to Lo while the clock signal CLK1 is Hi. Thereafter, the clock signal CLK1 transits from Hi to Lo. At that time, the node A1-1 and the node N1 are charged. However, depending on the transistor variation of the p-type MOS transistor, the node N1 first reaches the threshold voltage Vtn of the n-type MOS transistor. At this time, a through current flows through the node N2, a Glitch is generated at the node N2, the Glitch is propagated to the node N7, and the output terminal Q malfunctions.

  However, in the circuit of FIG. 15, since the source of the p-type MOS transistor 13A is connected to the node A1-1, the current characteristic between the drain and source of the p-type MOS transistor 13A is p-type MOS transistor 13A. The voltage between the drain and the source exhibits a linear characteristic with respect to the voltage between the drain and the source until the voltage near the threshold voltage Vtp. Further, when the substrate voltage and the source voltage of the p-type MOS transistor 13A are compared, the substrate voltage is higher, so that it behaves as a very high resistance element. For this reason, the node A1-1 is charged first, and then the node N1 starts to be charged. Therefore, after the gate voltage of the n-type MOS transistor 1E-1 becomes lower than the threshold voltage of the n-type MOS transistor, the gate voltage of the n-type MOS transistor 1E-2 tends to become higher than the threshold voltage. The through current hardly flows, and Glitch of the node N7 does not occur. Further, in FIG. 15, the gate of the p-type MOS transistor 13B2 and the gate of the p-type MOS transistor 13C2 are connected to the clock signal CLK1.

  Therefore, in the circuit of FIG. 12, the discharge of the node N2 is not started unless the node A1-2 becomes equal to or higher than the voltage (VDD-Vtp). However, in FIG. 15, the clock signal CLK1 is changed to the voltage (VDD-Vtp). ) Since the node N2 is discharged, there is an advantage that the operation of the node N2 can be performed at a higher speed than in FIG.

  As described above, the source of the p-type MOS transistor 13B2 is connected to the power supply, the drain of the p-type MOS transistor 13B2 is connected to the source of the p-type MOS transistor 13B1, and the drain of the first p-type MOS transistor 13B1 is connected to the node N2. The gate of the second p-type MOS transistor 13B2 is connected to the clock signal CLK1, the gate of the p-type MOS transistor 13B1 is connected to the output of the inverting circuit INV13 of the node N1, and the source of the p-type MOS transistor 13C2 is connected to the power supply. The drain of the p-type MOS transistor 13C1 is connected to the source of the p-type MOS transistor 13C1, the drain of the p-type MOS transistor 13C1 is connected to the node N4, and the gate of the p-type MOS transistor 13C2 is connected to the clock signal CLK1, p-type MOS. Dora By connecting the gate of the register 13C1 to the output of the inverting circuit INV13 of the node N1, and further connecting the source of the p-type MOS transistor 13A to the node A1-1, the charging order of the node A1-1 and the node N1; Since the order of charging between the nodes N1 and N2 and the order of charging between the nodes N1 and N4 are uniquely determined regardless of the manufacturing variation of the device size of the p-type MOS transistor, a more robust circuit configuration can be obtained. realizable.

  In the above description, the source of the p-type MOS transistor 13A is connected to the node A1-1. However, the source of the p-type MOS transistor 13A is connected to the drain of another p-type MOS transistor, and the source is used as a power source. A similar effect is exhibited even in a configuration in which the gates are connected to the output of the inverting circuit of the node A1-1. In other words, the intent of the present invention is that the order of charging between the node A1-1 and the node N1, the order of charging between the nodes N1 and N2, and the order of charging between the nodes N1 and N4 are determined according to the device of the p-type MOS transistor. Any circuit configuration may be used as long as the circuit configuration is uniquely determined without being influenced by manufacturing variations in size. The circuit configuration is possible by a combination of various circuits, but they do not depart from the present invention.

(Embodiment 6)
FIG. 19 shows another circuit diagram of the dynamic circuits 1C and 1D in the multi-input flip-flop of FIG.

  19 differs from the dynamic circuits 1C and 1D of FIG. 1 in that the first and second p-type MOS transistors A13 and N14A are arranged instead of the p-type MOS transistor Tr9 for charging the node N3. In the p-type MOS transistor N14A, the clock signal CLK3 is input to the gate, the source is connected to the power supply, and the drain is the node A2-2 (that is, the common source of the third n-type transistor group A20 to AK). And the source and drain of the other p-type MOS transistor A13 are connected to the node N3 and the node A2-2 (that is, the common drain and the common source of the third n-type transistor group A20 to AK), respectively. This is the point. Further, in FIG. 1, the clock signal CLK4 is connected to the gate terminal of the transistor Tr14 of the dynamic circuit 1D. However, in FIG. 19, the output of the inverting circuit IN14 is connected to the gate terminal of the transistor Tr14.

  As an advantage of such a circuit configuration, when the same phase as the clock signal CLK3 is input to the clock signal CLK4, that is, when driven by only the clock signal CLK3 as shown in FIG. It is a point that can work with. This is because in the circuit configuration of FIG. 1, when the clock signal CLK4 and the clock signal CLK3 are input in the same phase, the power supply voltage is a low voltage near the threshold voltage of the n-type MOS transistor (for example, the threshold value of the n-type MOS transistor). When the voltage is 0.3 V and the power supply voltage value is 0.5 V), when the clock signal CLK3 changes from Lo to Hi, the time at which the node N3 is discharged is much slower than the gate terminal of the transistor Tr14. Although the transistor Tr13 is cut off and the node N4 becomes Hi (that is, any of the terminals S [1] to S [N] and the terminal SE is Hi), the node N4 is It becomes Lo without becoming Hi.

  However, in the configuration of FIG. 19, when the clock signal CLK3 transitions from Lo to Hi, the nodes N3 and A2-2 start discharging at the same time, and when the node N14A falls below the switching level of the inverting circuit IN14, the gate of the transistor Tr14 The voltage rises. That is, since the node N3 becomes equal to or lower than the threshold voltage of the n-type MOS transistor Tr13 before the gate of the transistor Tr14 becomes Hi, a through current through the transistors Tr13 and Tr14 hardly occurs at the node N4. Low voltage operation is more stable than the circuit configuration.

  Further, when the clock signal CLK3 transitions from Hi to Lo, the current characteristic between the drain and the source of the p-type MOS transistor A13 indicates that the voltage between the drain and the source of the p-type MOS transistor A13 is close to the threshold value Vtp. Shows a linear characteristic with respect to the voltage between the drain and the source. Further, when the substrate voltage and the source voltage of the p-type MOS transistor A13 are compared, the substrate voltage is higher, so that it behaves as a very high resistance element. The node N3 is not charged unless the potential of the node A2-2 becomes equal to or higher than the threshold voltage of the p-type MOS transistor A13. That is, the transistor Tr13 is not turned on until the gate of the transistor Tr14 is lowered to some extent. Further, since the node N4 is charged by the clock signal CLK3, a glitch caused by turning on the transistor Tr13 is less likely to occur in the potential of the node N4. As a result, a malfunction related to the dynamic circuits A1 and 1D is unlikely to occur. Become.

(Embodiment 7)
FIG. 20 shows an application example of FIG.

  In FIG. 11, the input data is a flip-flop with multi-input selection divided into two sets, but in FIG. 20, the multi-input selection composed of dynamic circuits 1 </ b> A to 1 </ b> D and A <b> 1 is made by combining the transistors of each output circuit 1 </ b> E. The NAND logic for the function and the output of the multi-input selection function including the dynamic circuits 1A ′ to 1D ′ and A1 ′ is configured.

  Specifically, the other P-type MOS transistor Tr20 is arranged in common with the source and drain of one p-type MOS transistor Tr20, and the other n-type MOS transistor Tr21 is arranged in series with one n-type MOS transistor Tr21. is doing. Further, in the holding circuit 50 including the first inverter circuit INV15 connected to the drain of the p-type MOS transistor Tr20 and the second inverter INV16 having the output of the inverter INV15 as an input, the second inverter INV16 An n-type MOS transistor 16A in which the second output node N2 of the dynamic circuits 1A to 1D and A1 is connected to the gate is arranged in a single stage between the P-type MOS transistor, Tr60, and N-type MOS transistor Tr61. In addition to this (this configuration is the same in FIG. 11), in FIG. 20, the n-type MOS is further connected to the gate of the second output node N2 ′ of the dynamic circuits 1A ′ to 1D ′ and A1 ′. Transistor 16B is arranged in one stage in series. Thereby, the high speed of the holding circuit 50 is maintained. These two-stage n-type MOS transistors may be disposed between the ground and the n-type MOS transistor Tr61 constituting the second inverter circuit INV16.

  In this embodiment, an example of NAND logic has been described. However, the present invention is not limited to this example, and it goes without saying that various composite logics can be generated. Furthermore, by replacing the dynamic logic part related to the dynamic circuits 1A and 1A 'with various logics, a flip-flop circuit having various composite logic functions can be configured. Furthermore, one signal selected from a plurality of input signals is branched into a plurality of signals, and each of them is connected to a NAND circuit, a NOR circuit, an EXOR circuit, etc., and is different for the selected one signal. A configuration may be employed in which a plurality of signals subjected to logical operations are output. Further, adding a MOS transistor to the transistor Tr20 or the transistor Tr21 and connecting the gate terminal thereof to the output of another multi-input dynamic circuit does not depart from the present invention.

  FIG. 21 shows another application example of FIG. 11, in which the source and drain of the transistor Tr21 in each output circuit are connected in common.

(Embodiment 8)
FIG. 22 shows another application example of FIG. 11 in which only the scan input circuit is placed in the dynamic circuits 1A ′ to 1D ′ and A1 ′.

  The dynamic circuits 1A ′ to 1D ′, A1 ′, 17B, and 17C share the holding circuit unit 17E and the output unit of the output terminal Q with a flip-flop having a multi-input selection function that includes the dynamic circuits 1A to 1D and A1. It is a static type flip-flop. Furthermore, the difference from FIG. 11 is that the gate of the n-type MOS transistor 17D is connected to the inverted output of the scan enable signal SE.

  By adopting such a circuit configuration, when the scan enable signal is activated, the transistors Tr22 and Tr20 are cut off and only the circuit elements 17B and 17C operate. The advantage of this circuit is that the capacity of the node N1 can be reduced, and in the normal path, a high speed is achieved by using a dynamic type flip-flop, and in the scan path, a scan input is achieved by using a static type flip-flop. Hold time is shortened, which is effective in securing a margin for the scan shift operation.

  It goes without saying that flip-flop circuits having various logic functions can be configured by combining the output circuit portion of the dynamic circuit and the output portion of the static circuit with the output circuit portion 17F. In the present invention, as described above, the advantages of the dynamic circuit and the static circuit can be properly used in accordance with the input signal function and specifications.

  Although the eight embodiments have been described above, replacing a part of the circuit configuration of the semiconductor integrated circuit in one of the embodiments with a part of the circuit configuration of any of the other seven embodiments is not appropriate. Easy for contractors. For example, the dynamic circuit 1B in FIG. 8 may be replaced with the dynamic circuit 1B in FIG.

  As described above, according to the present invention, it is possible to satisfactorily hold the output signal of the holding circuit at the previous value even when none of the selection signals is activated and all the data is selected. Therefore, it is useful as a dynamic flip-flop circuit with a data selection function.

  Further, according to the present invention, when the input data already matches the value of the output signal from the holding circuit, the operation of at least a part of the dynamic flip-flop circuit can be forcibly stopped. Therefore, it is suitable to apply to a semiconductor integrated circuit or the like that suppresses the operation and further reduces power consumption.

It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 1 of this invention. It is a figure which shows the outline of the layout structure of the principal part of the semiconductor integrated circuit. (A) is a figure which shows the principal part structure of the conventional semiconductor integrated circuit, (b) is a figure which shows the example of 1 proposal which eliminates the fault of the semiconductor integrated circuit. It is a figure which shows the operation | movement timing chart of the semiconductor integrated circuit of Embodiment 1 of this invention. It is a figure which shows the internal structure of the output circuit with which the semiconductor integrated circuit of Embodiment 2 of this invention is equipped. It is a figure which shows the internal structure of the generation circuit of the clock given to the output circuit. It is a figure which shows the operation | movement timing chart of the output circuit and a clock generation circuit. It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 3 of this invention. FIG. 10 is a diagram showing a modification of the semiconductor integrated circuit shown in FIG. 1. FIG. 10 is a diagram showing a layout configuration of a main part of the semiconductor integrated circuit shown in FIG. 9. FIG. 10 is a diagram showing a further modification of the semiconductor integrated circuit shown in FIG. 9. It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 4 of this invention. 10 is a timing chart of each node in an input pattern of various signals of the semiconductor integrated circuit of Embodiment 4. FIG. It is a figure which shows the timing chart of each node in another input pattern of the semiconductor integrated circuit of Embodiment 4. It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 5 of this invention. 10 is a timing chart of each node in an input pattern of various signals of the semiconductor integrated circuit of Embodiment 5. FIG. It is a figure which shows the timing chart of each node in another input pattern of the semiconductor integrated circuit of Embodiment 5. It is a figure which shows the timing chart of each node in another input pattern of the semiconductor integrated circuit of the same Embodiment 5. FIG. It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 6 of this invention. It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 7 of this invention. It is a figure which shows the structure of the modification of the semiconductor integrated circuit of Embodiment 7 of this invention. It is a figure which shows the structure of the semiconductor integrated circuit of Embodiment 8 of this invention.

1A NOR type first dynamic circuit 2A NOR type first dynamic circuit 1B NAND type second dynamic circuit 1C NOR type third dynamic circuit (non-selection state detection circuit)
2C NOR type third dynamic circuit (coincidence detection circuit)
1D NAND-type fourth dynamic circuit 1E output circuit Tr20 first p-type transistor Tr21 first n-type transistor Tr22 second n-type transistor 1F holding circuit IN5 first inverter circuit IN6 second inverter circuit Tr27 first P-type transistor Tr28 first n-type transistor Tr29 second n-type transistor 1G output circuit 70 differential circuit 71 OR circuit Tr36 n-type transistor (control transistor)
Tr37 n-type transistor (resistance element)
1H clock generation circuit (signal generation circuit)
75 Short pulse generation circuit NAND1 NAND circuit

Claims (31)

In a semiconductor integrated circuit that receives a clock, a plurality of data, and a plurality of selection signals for selecting each data, and outputs one data selected by the selection signal to a holding circuit when the clock transitions,
A non-selection state detection circuit that detects a state in which all of the plurality of selection signals does not select any of the plurality of data;
When a state in which all of the plurality of selection signals do not select any of the plurality of data is detected in the non-selection state detection circuit, the change in the data selected last time is prevented, and the output data of the holding circuit The semiconductor integrated circuit characterized by holding. The first clock and the plurality of data are input, and the charge is the first in the first period which is one of the period from the rising edge to the falling edge and the period from the falling edge to the rising edge of the first clock. When the output node is charged and the values of the plurality of data all match in the second period, which is the other period, the charge of the first output node is held, while at least one of the plurality of data A NOR-type first dynamic circuit that discharges the charge of the first output node when the two do not match,
The second clock and the signal of the first output node of the first dynamic circuit are input, and the charge of the first output node of the first dynamic circuit is the first period or the second period of the second clock. Is discharged when the first output node is discharged, while the second output node is discharged when the first output node is held. In a semiconductor integrated circuit comprising a dynamic circuit,
A third clock and a plurality of selection signals for selecting each of the plurality of data are input, and a charge is charged in a third output node in a first period of the third clock, and in a second period, A NOR-type third dynamic circuit that holds the charge of the third output node when all of the plurality of selection signals do not select any of the plurality of data;
The fourth clock and the signal of the third output node of the third dynamic circuit are input, and the charge of the third output node of the third dynamic circuit in the first period or the second period of the fourth clock. A NAND-type fourth dynamic circuit that discharges the charge of the fourth output node when is held,
The second dynamic circuit receives a signal from the fourth output node of the fourth dynamic circuit, and when the electric charge at the fourth output node is discharged, the first output of the first dynamic circuit. A semiconductor integrated circuit characterized by holding the charge of the second output node even when the charge of the node is held. 3. The semiconductor integrated circuit according to claim 2, wherein
The NOR-type third dynamic circuit and the NAND-type fourth dynamic circuit are closer to the NAND-type second dynamic circuit than the NOR-type first dynamic circuit in terms of physical arrangement. A semiconductor integrated circuit. 3. The semiconductor integrated circuit according to claim 2, wherein
The NOR type third dynamic circuit and the NAND type fourth dynamic circuit operate at a higher speed than the NOR type first dynamic circuit. 3. The semiconductor integrated circuit according to claim 2, wherein
The NOR-type third dynamic circuit and the NAND-type fourth dynamic circuit have a higher supply voltage than the NOR-type first dynamic circuit. 3. The semiconductor integrated circuit according to claim 2, wherein
The NOR type third dynamic circuit and the NAND type fourth dynamic circuit are located farther from the element isolation region formed on the semiconductor substrate than the NOR type first dynamic circuit in terms of physical arrangement. A semiconductor integrated circuit. 3. The semiconductor integrated circuit according to claim 2, wherein
An output circuit for receiving the selected data and outputting the selected data;
The output circuit is
A NOR circuit that receives an output of the NOR-type first dynamic circuit and an inverted output of the output of the NAND-type second dynamic circuit;
A first n-type transistor in which an output of the NOR circuit is input to a gate;
A first p-type transistor having an output of the NAND-type second dynamic circuit input to a gate;
A drain is connected to the first n-type transistor and the first p-type transistor. A semiconductor integrated circuit, wherein: The semiconductor integrated circuit according to claim 7, wherein
The output circuit is
Furthermore, it has a second n-type transistor in which the output of the NAND-type fourth dynamic circuit is input to the gate,
A semiconductor integrated circuit, wherein a drain of the second n-type transistor is connected to a source of the first n-type transistor. The semiconductor integrated circuit according to claim 7, wherein
A holding circuit for holding the selection data is connected to a drain of the first n-type transistor and a drain of the first p-type transistor which are two output terminals of the output circuit. Integrated circuit. The semiconductor integrated circuit according to claim 9, wherein
The holding circuit is
A first inverter circuit in which a drain of the first p-type transistor of the output circuit is connected to an input side;
A second inverter circuit in which an output of the first inverter circuit is input and a first n-type transistor and a first P-type transistor are connected in series;
A second n-type transistor receiving the output of the NAND-type second dynamic circuit,
The second n-type transistor is disposed between the n-type transistor and the p-type transistor of the second inverter circuit, or between the n-type transistor of the second inverter circuit and the ground. A semiconductor integrated circuit. 3. The semiconductor integrated circuit according to claim 2, wherein
An output circuit for receiving the selected data and outputting the selected data;
The output circuit is
A differential circuit having a differential input terminal and a differential output terminal;
An OR circuit to which an output of the NOR-type first dynamic circuit and an inverted output of the output of the NAND-type second dynamic circuit are input;
The output of the OR circuit is input to one of the differential input terminals of the differential circuit,
The semiconductor integrated circuit, wherein the output of the second NAND dynamic circuit is input to the other differential input terminal of the differential circuit. The semiconductor integrated circuit according to claim 11, wherein
The differential circuit includes a control transistor that enables a differential amplification operation of the differential circuit,
The control transistor is
A semiconductor integrated circuit comprising: an n-type transistor having a drain connected to a source of the differential circuit, a source grounded, and a gate receiving a control signal. The semiconductor integrated circuit according to claim 12, wherein
A resistance element is connected in parallel to the control transistor. A semiconductor integrated circuit, wherein: The semiconductor integrated circuit according to claim 12, wherein
A signal generation circuit for generating a control signal to be applied to the gate of the control transistor;
The signal generation circuit includes:
A short pulse generation circuit for generating a short pulse from a clock signal;
A NAND circuit to which the short pulse and the output of the NAND type fourth dynamic circuit are input,
An output of the NAND circuit is given to the gate of the control transistor as the control signal. A semiconductor integrated circuit, wherein: 3. The semiconductor integrated circuit according to claim 2, wherein
In the NOR type third dynamic circuit and the NAND type fourth dynamic circuit, the threshold voltage of the transistors constituting them is more than the threshold voltage of the transistors constituting the NOR type first dynamic circuit. A semiconductor integrated circuit characterized by being low. 3. The semiconductor integrated circuit according to claim 2, wherein
The semiconductor integrated circuit characterized in that the first, second, third and fourth clock signals are shared by one clock signal. 3. The semiconductor integrated circuit according to claim 2, wherein
An inversion node of the third output node of the third dynamic circuit is connected to the second dynamic circuit,
The second dynamic circuit includes:
When the charge of the inversion node of the third output node is charged and the charge of the fourth output node of the fourth dynamic circuit is held, the second output node of the second dynamic circuit Discharging the charge,
The semiconductor integrated circuit, wherein the charge of the inversion node of the third output node is held and the charge of the second output node is held when the charge of the fourth output node is discharged. . 3. The semiconductor integrated circuit according to claim 2, wherein
The third dynamic circuit includes a first n-type transistor to which the third clock signal is input to a gate, and a second n-type transistor having a source connected in common to the drain of the first n-type transistor. A group, and a third n-type transistor group having a source connected in common to the drain of the second n-type transistor group,
Of the gate potentials of the second n-type transistor group, at least one gate potential is set to the power supply potential, and the other gate potentials are set to the ground potential.
The gates of the third n-type transistor group are each connected to one of the plurality of selection signals, and the drains of the third n-type transistor group are commonly connected to the third output node. ,
An inversion node of the third output node of the third dynamic circuit is connected to the second dynamic circuit,
In addition to the third output node, the fourth dynamic circuit is connected to an inversion node of the drain connected in common to the second n-type transistor group,
The second dynamic circuit includes:
When the charge of the inversion node of the third output node is charged and the charge of the fourth output node of the fourth dynamic circuit is held, the second output node of the second dynamic circuit Discharging the charge,
The semiconductor integrated circuit, wherein the charge of the inversion node of the third output node is held and the charge of the second output node is held when the charge of the fourth output node is discharged. . 3. The semiconductor integrated circuit according to claim 2, wherein
The third dynamic circuit includes a first n-type transistor to which the third clock signal is input to a gate, and a third n-type transistor group having sources connected in common.
The gates of the third n-type transistor group are each connected to one of the plurality of selection signals, and the drains of the third n-type transistor group are commonly connected to the third output node. ,
The fourth dynamic circuit includes:
When the charge of the third output node is held, the charge of the fourth output node is discharged,
The charge of the fourth output node is held when the charge of the third output node is discharged. The semiconductor integrated circuit, wherein The semiconductor integrated circuit according to claim 18 or 19 ,
In the third dynamic circuit, the third clock signal is input to the gate, the drain is connected to the source of the third n-type transistor group, and the third clock signal is input to the third dynamic circuit. And a second p-type transistor having a gate input, a source connected to a drain of the third n-type transistor group, and a drain connected to a source of the third n-type transistor group. A semiconductor integrated circuit. 3. The semiconductor integrated circuit according to claim 2, wherein
The semiconductor integrated circuit, wherein the charge charged in the first output node is supplied from the third output node. 3. The semiconductor integrated circuit according to claim 2, wherein
The semiconductor integrated circuit characterized in that the charge charged in the second output node is supplied from the first output node. 3. The semiconductor integrated circuit according to claim 2, wherein
The semiconductor integrated circuit, wherein the charge charged in the fourth output node is supplied from the first output node. 3. The semiconductor integrated circuit according to claim 2, wherein
The first dynamic circuit includes a first p-type transistor having a gate connected to an inverted output of the first output node of the first dynamic circuit, and a second having a gate connected to the first clock signal. P-type transistor
The first p-type transistor and the second p-type transistor are connected in series, the source of one p-type transistor is connected to a power supply, and the drain of the other p-type transistor is the fourth output node or the A semiconductor integrated circuit connected to the second output node. The semiconductor integrated circuit according to claim 18 , wherein
The potential of at least one of the gates of the second n-type transistor group is connected to a power supply potential via a potential setting transistor, and the second n-type transistor group and the potential setting transistor are the same standard. A semiconductor integrated circuit characterized by being arranged in a cell. The semiconductor integrated circuit according to claim 25, wherein
The potential setting transistor is the p-type transistor having a drain connected to at least one gate of the gates of the second n-type transistor group;
A semiconductor integrated circuit comprising an n-type transistor having a source grounded and a drain and a gate connected to the potential setting transistor in the same standard cell. A semiconductor integrated circuit comprising two semiconductor integrated circuits according to claim 7,
The sources and drains of each first n-type transistor are connected in common,
A semiconductor integrated circuit, wherein the sources and drains of each first p-type transistor are connected in common. A semiconductor integrated circuit comprising two semiconductor integrated circuits according to claim 7,
A logic is formed by using each of the output circuits. A semiconductor integrated circuit, wherein: The semiconductor integrated circuit according to claim 28, wherein
A first inverter circuit in which drains of the first p-type transistors of each of the output circuits are connected in common, and a second inverter circuit to which an output of the first inverter circuit is input; Holding circuit is configured,
The corresponding first p-type transistor and the gate are made common between the n-type transistor and the p-type transistor constituting the second inverter circuit or between the n-type transistor and the ground, respectively. A semiconductor integrated circuit, wherein n-type transistors are arranged in series. A semiconductor integrated circuit comprising the semiconductor integrated circuit according to claim 7 and a static flip-flop,
The output circuit receives the output of the static flip-flop, and outputs either the selected data or the output of the static flip-flop. The semiconductor integrated circuit according to claim 30, wherein
Scan test data is input to the static flip-flop. A semiconductor integrated circuit, wherein:

Images