阅读说明：本技术 降低芯片逻辑翻转率的方法及结构 (Method and structure for reducing chip logic turnover rate ) 是由 何文俊 郭凯 于 2019-08-19 设计创作，主要内容包括：发明提供了一种降低芯片逻辑翻转率的方法及结构,N个寄存器中存储的数据不参与到传递过程中多级寄存器的翻转,N为大于等于1的自然数,而是当所述反向标识的值为预设标识的值时,直接将N个寄存器中存储的数据与N位初始信号进行异或运算,从而得到预期信号,整体上有效的降低了逻辑翻转率,从而降低了系统功耗。(The invention provides a method and a structure for reducing the logic turnover rate of a chip, wherein data stored in N registers does not participate in the turnover of a multi-stage register in the transmission process, N is a natural number which is more than or equal to 1, and when the value of a reverse identifier is the value of a preset identifier, the data stored in the N registers and an N-bit initial signal are directly subjected to XOR operation to obtain an expected signal, so that the logic turnover rate is effectively reduced on the whole, and the power consumption of a system is reduced.)

1. A method for reducing the logic turnover rate of a chip, which is used for reducing the logic turnover rate of signals in the chip, and is characterized by comprising the following steps:

acquiring N initial signals and N expected signals of the initial signals;

for the ith initial signal in the N initial signals, executing the following steps:

recording the value of the reverse identification as a preset identification value, and storing an exclusive OR result of an ith bit initial signal and an ith bit expected signal in an ith register;

when the N-bit initial signal is needed, acquiring the ith-bit expected signal according to the data stored in the ith register and the ith-bit initial signal;

wherein N and i are natural numbers of 1 or more.

2. The method for reducing chip logic flip rate according to claim 1, wherein when the flip rate is greater than or equal to a preset flip rate, the value of the reverse flag is recorded as the preset flag value.

3. The method for reducing the logic flip rate of a chip according to claim 2, wherein the exclusive or results in the N registers are summed, and when the sum is greater than or equal to (1/preset flip rate), the flip rate is greater than or equal to the preset flip rate.

4. The method for reducing chip logic flip rate according to claim 2, wherein the predetermined flag value is 1 or 0.

5. A structure for reducing the logic turnover rate of a chip, which is used for reducing the logic turnover rate of N-bit signals in the chip, is characterized by comprising: n exclusive-OR gates, at least one adder and N registers;

for the ith initial signal in the N-bit signals, respectively executing the following operations:

inputting an ith bit initial signal and an ith bit expected signal corresponding to the ith bit initial signal into an ith exclusive-OR gate to obtain an ith exclusive-OR result, and storing the ith exclusive-OR result in an ith register;

inputting the N XOR results into the at least one adder to obtain a sum of the XOR results;

wherein N and i are both natural numbers greater than or equal to 1.

6. The structure for reducing the flip rate of chip logic of claim 5, further comprising an XOR summation register for storing the XOR result and value.

7. The structure for reducing the logic flip rate of a chip of claim 5, further comprising an inverse flag register for storing the value of the inverse flag.

Technical Field

The invention belongs to the field of integrated circuit chip design, and relates to a method and a structure for reducing the logic turnover rate of a chip.

Background

In chip design, all logic flips are transmitted backwards step by step; the logic turnover rate of the system is high, the middle stage is partially turned over without load, and in order to obtain correct output, the system is always turned over in the air, so that the waste of dynamic power consumption is caused; and if the design requirement of the intermediate register is excessive, the dynamic power consumption of the system is multiplied.

Many products have very strict requirements on the power consumption of a core chip, and the large power consumption not only can cause the acceleration of the power consumption and depreciation of the chip, but also can cause the problems of heating, the rise of the heat dissipation cost of the chip and the like; for mobile end products, power consumption is also one of the most important performance indexes; in the prior art, the redundant turning part of the multi-bit signal in the pipeline structure is not processed, and the consideration of power consumption is under optimization in design.

Accordingly, a method and structure for reducing the logic flip rate of a chip are needed to reduce the dynamic power consumption of the system.

Disclosure of Invention

Referring to fig. 1, fig. 1 is a circuit diagram illustrating a process of reducing a logic slew rate. In the figure, 4-bit low level signals are inverted into 4-bit high level signals, then M-level register running water is transferred, and finally the M-level register running water is sent to a load end for use. Wherein M is a natural number greater than 1. As can be seen from FIG. 1, this design results in premature logic rollover and stage-by-stage transfer rollover rates of the intermediate stage register pipeline. The 4-bit initial signal is 0000, the expected signal corresponding to it is 1111, i.e. the inversion rate is 1111 from 0000, the inversion rate is 100%, 4 inversions are needed, and assuming that the 4-level register (in practical cases, there are more levels) is passed and then used, the total number of inversions is 20.

The invention aims to provide a method and a structure for reducing the logic turnover rate of a multi-bit signal from the aspect of chip design so as to reduce the dynamic power consumption of a chip on the premise of meeting the design function of the chip.

The invention provides a method for reducing the logic turnover rate of a chip, which is used for reducing the logic turnover rate of signals in the chip and comprises the following steps:

acquiring N initial signals and N expected signals of the initial signals;

for the ith initial signal in the N initial signals, executing the following steps:

recording the value of the reverse identification as a preset identification value, and storing an exclusive OR result of an ith bit initial signal and an ith bit expected signal in an ith register;

when the N-bit initial signal is needed, acquiring the ith-bit expected signal according to the data stored in the ith register and the ith-bit initial signal;

wherein N and i are natural numbers of 1 or more.

Preferably, in the method for reducing the chip logic inversion rate, when the inversion rate is greater than or equal to a preset inversion rate, the value of the reverse flag is recorded as the preset flag value.

Preferably, in the method for reducing the chip logic inversion rate, the xor results in the N registers are summed, and when the sum is greater than or equal to (1/preset inversion rate), the inversion rate is greater than or equal to the preset inversion rate.

Preferably, in the method for reducing the chip logic flip rate, the preset identification value is 1 or 0.

The invention also provides a structure for reducing the logic turnover rate of the chip, which is used for reducing the logic turnover rate of the N-bit signal in the chip and comprises the following steps: n exclusive-OR gates, at least one adder and N registers;

for the ith initial signal in the N-bit signals, respectively executing the following operations:

inputting an ith bit initial signal and an ith bit expected signal corresponding to the ith bit initial signal into an ith exclusive-OR gate to obtain an ith exclusive-OR result, and storing the ith exclusive-OR result in an ith register;

inputting the N XOR results into the at least one adder to obtain a sum of the XOR results;

wherein N and i are both natural numbers greater than or equal to 1.

Preferably, in the above structure for reducing the chip logic flip rate, an xor summation register is further included for storing the xor result and the value.

Preferably, in the structure for reducing the chip logic flip rate, a reverse flag register is further included for storing a value of the reverse flag.

The invention provides a method and a structure for reducing the logic turnover rate of a chip, wherein data stored in N registers does not participate in the turnover of a multi-stage register in the transmission process, N is a natural number which is more than or equal to 1, and when the value of a reverse identifier is the value of a preset identifier, the data stored in the N registers and an N-bit initial signal are directly subjected to XOR operation, so that an expected signal is obtained, the logic turnover rate is effectively reduced on the whole, and the power consumption of a system is reduced.

Drawings

FIG. 1 is a schematic diagram of a circuit without a reduced logic slew rate process;

FIG. 2 is a flowchart illustrating a method for reducing the logic flip rate of a chip according to an embodiment of the present invention;

FIG. 3 is a diagram illustrating a structure for reducing the logic flip rate of a chip according to an embodiment of the present disclosure;

FIG. 4 is a diagram illustrating the reduction of the inversion rate of a 4-bit initial signal according to the present invention.

Detailed Description

In the following description, numerous technical details are set forth in order to provide a better understanding of the present invention. However, it will be understood by those skilled in the art that the claimed embodiments of the present invention may be practiced without these specific details and with various changes and modifications based on the following embodiments.

In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

An embodiment of the present invention provides a method for reducing a chip logic slew rate, which is used to reduce a logic slew rate of a signal in a chip, and specifically, as shown in fig. 2, fig. 2 is a flowchart of the method for reducing the chip logic slew rate in an embodiment of the present invention, and includes the following steps: acquiring an N-bit initial signal and an N-bit expected signal of the initial signal, as in step S202 in fig. 2; for the ith initial signal in the N-bit initial signals, marking the value of the reverse flag as a preset flag value, and storing the xor result of the ith initial signal and the ith expected signal in the ith register, as in step S204 in fig. 2; and when the N-bit initial signal is needed, acquiring the i-th bit expected signal according to the data stored in the i-th register and the i-th bit initial signal, as in step S206 in fig. 2, where N and i are natural numbers greater than or equal to 1.

Specifically, when the turnover rate is greater than or equal to a preset turnover rate, the value of the reverse identifier is recorded as the preset identifier value. In one embodiment of the present specification, the preset identification value is 50%, but in other embodiments of the present invention, the preset identification value may be any other value between 0 and 1, for example, 25%. According to the habit of the engineer or the requirement of the project engineering, the preset identification value may be marked as 0 or 1, which is not limited herein.

Further, the XOR results in the N registers are summed, and when the sum is greater than or equal to (1/preset turnover rate), the logic turnover rate of the current chip is greater than or equal to the preset turnover rate.

Still another embodiment of the present invention further provides a structure for reducing a chip logic flip rate, configured to reduce a logic flip rate of an N-bit signal in a chip, where N is a natural number greater than or equal to 1, as shown in fig. 3, and fig. 3 is a schematic diagram of a structure for reducing a chip logic flip rate in an embodiment of the present disclosure, where the structure includes: n exclusive or gates, at least one adder, N registers, and an exclusive or sum register. Aiming at the ith initial signal in the N-bit signals, respectively executing the following steps: firstly, inputting an ith bit initial signal, an ith bit corresponding to the ith bit initial signal and a signal of the ith bit initial signal into an ith exclusive-OR gate to obtain an ith exclusive-OR result, and storing the ith exclusive-OR result in an ith register, wherein i is a natural number which is more than or equal to 1; inputting the N XOR results into the at least one adder to obtain a sum of XOR results, and storing the sum of XOR results in the XOR summing register.

Furthermore, the structure for reducing the chip logic turnover rate further comprises a reverse identification register for storing a reverse identification value. Specifically, in an embodiment of the present disclosure, when the turnover rate is greater than or equal to a preset turnover rate, the value of the reverse indicator is recorded as the preset indicator value.

Furthermore, the exclusive or results in the N registers are summed, and when the sum is greater than or equal to (1/preset turnover rate), the turnover rate is greater than or equal to the preset turnover rate.

The following description will be given with reference to a specific example. Fig. 4 is a schematic diagram illustrating the reduction of the inversion rate of a 4-bit initial signal according to the present invention. The 4-bit initial signal is 0000, the expected signal corresponding to the 4-bit initial signal is 1111, namely 0000 is turned into 1111, the preset turning rate is 50%, 4 exclusive or gates and a 4-bit adder are adopted for judgment, at the moment, 4 exclusive or results are all 1, the turning rate is 100%, the turning rate is greater than 50% of the preset turning rate, the value of the reverse identification is set to 1, 4 registers are used for storing the exclusive or results, 4-bit data are not turned, and the 4-level registers are not turned in the process of transferring the 4-level registers until a load end needs to use the 4-bit initial signal to participate in logic operation.

At this time, the value of the reverse flag is recognized first, and at this time, the value of the reverse flag is 1, and then the xor result (0000) stored in the 4 registers is taken out, and is subjected to xor operation with the 4-bit initial signal before inversion (i.e., 0000), so that 1111 is the expected signal is obtained.

In the process, although N registers are newly added, multi-bit logic flip of intermediate 4-level register pipelining is omitted, and the logic flip rate can be effectively reduced on the whole, so that the power consumption of the system is reduced.

In the method and the structure for reducing the logic turnover rate of the chip, provided by the invention, the data stored in the N registers do not participate in the turnover of the multi-stage registers in the transmission process, N is a natural number which is more than or equal to 1, and when the value of the reverse identification is the value of the preset identification, the data stored in the N registers and the N-bit initial signal are directly subjected to XOR operation, so that an expected signal is obtained, the logic turnover rate is effectively reduced on the whole, and the power consumption of a system is reduced.

Various other modifications and changes may be made by those skilled in the art based on the above-described technical solutions and concepts, and all such modifications and changes should fall within the scope of the claims of the present invention.