Quasi-full maximum distance random testing method

Abstract

The invention relates to a quasi-full maximum distance random testing method. The method comprises the following testing steps: reading the primary input number n of a circuit from a text, and specifying an initial seed n-bit binary code t0 to generate a desired testing code number 2m; generating a subscript odd testing code set which is accordant with a maximum total Descartes distance according to the primary input number n; further acquiring a subscript even testing code set according to the rule of the maximum Hamming distance on the basis of the obtained subscript odd testing code set to form a middle testing set; performing exclusive-or operation on the middle testing set and a specified testing seed t0 to obtain a final complete testing set. By adopting the quasi-full maximum distance random testing method, the problems of huge resource consumption and time consumption for achieving an ideal fault covering rate since random testing codes are not excessive in the conventional random testing algorithm under the condition that the circuit scale increases sharply at present and integrated circuits develop rapidly are solved. The method has the characteristics that the testing efficiency is increased, and the fault covering rate is increased.

Description

technical field

[0001] The invention relates to the field of computer hardware fault testing, in particular to the field of fixed fault diagnosis algorithms based on random testing, and proposes a quasi-complete maximum distance random testing method. Background technique

[0002] Basic knowledge introduction:

[0003] As a black-box testing technology that does not require internal information of the circuit under test, random testing avoids spending a lot of time and energy on analyzing the structure of the circuit under test, and does not require 100% fault coverage, such as only 90%-95% It has been widely used in practical industrial tests. Random testing is a well-known fixed fault diagnosis algorithm that has been used for many years. The random test method needs to randomly generate test cases for a circuit, that is, for a circuit with n For circuits with input variables, it will be randomly generated n A bit test pattern is loaded onto the circuit, and the ...

Examples

Embodiment 1

[0034] see figure 1 , the standard complete maximum distance random test method test steps are as follows:

[0035] Step 1, read the circuit's primary input number from the text n , specifying the initial seed n specified test seed in binary code t 0 , the number of test patterns expected to be generated is 2 m ;

[0036] Step 2, according to the number of primary inputs n , to generate a set of subscripted odd test codes that meet the maximum total Cartesian distance;

[0037] Step 3. On the basis of obtaining the subscript odd test code set, according to the rule of the maximum Hamming distance, further obtain the subscript even test code set to form an intermediate test set;

[0038] Step 4, intermediate test set and specified test seeds t 0 Perform XOR operation to obtain the final complete test set.

Embodiment 2

[0040] This embodiment is basically the same as Embodiment 1, and its special features are as follows:

[0041] Said step 2 includes:

[0042] Step 2.1, according to the number of primary inputs n , calculate the best distance as the closest n Even number of / 2 σ , to select candidate codes;

[0043] Step 2.2, according to the selected optimal distance σ , set the distance range range = σ ,number of failures fcount =0; random from n Bit binary test code selected from σ The ones place is juxtaposed to 1, and the generated Hamming distance from all 0 yards is σ The test code of the test code is used as the candidate code; if the Hamming distance between the candidate code and all the generated subscript odd test codes is within range Within the range, go to step 2.3; otherwise, fcount Add 1, repeat step 2.2; if fcount exceeds the specified threshold, then relax range scope: range =[ range lower limit -2, range cap+2];

[0044] Step 2.3, use the ca...

Embodiment 3

[0048] In this embodiment, the quasi-complete maximum distance random testing method is based on Microsoft visual studio 2010, and is implemented with C++ object-oriented language as the programming language.

[0049] figure 1 As shown, the complete maximum distance random test method in this paper includes the following steps, and its characteristics are:

[0050] Step 1, read the number of input terminals of the circuit from the text n , the target test set size is 2 m ;

[0051] Step 2, initialize each parameter, generate candidate test codes, and obtain all subscript odd test codes;

[0052] In step 2 mentioned include:

[0053] Step 2.1, use all "0" as the initial test seed t 0 , to calculate the optimal distance even number σ , which has σ The sum of the Cartesian distances between any test code of one "1" and all "0" and all "1" is the largest. when n ≠4 k +2; ( k , n?N )hour, σ =[ n / 2] (even number); when n =4 k +2; ( k , n?N ), σ...

technical field

[0001] The invention relates to the field of computer hardware fault testing, in particular to the field of fixed fault diagnosis algorithms based on random testing, and proposes a quasi-complete maximum distance random testing method. Background technique

[0002] Basic knowledge introduction:

[0003] As a black-box testing technology that does not require internal information of the circuit under test, random testing avoids spending a lot of time and energy on analyzing the structure of the circuit under test, and does not require 100% fault coverage, such as only 90%-95% It has been widely used in practical industrial tests. Random testing is a well-known fixed fault diagnosis algorithm that has been used for many years. The random test method needs to randomly generate test cases for a circuit, that is, for a circuit with n For circuits with input variables, it will be randomly generated n A bit test pattern is loaded onto the circuit, and the ...

Examples

Embodiment 1

[0034] see figure 1 , the standard complete maximum distance random test method test steps are as follows:

[0035] Step 1, read the circuit's primary input number from the text n , specifying the initial seed n specified test seed in binary code t 0 , the number of test patterns expected to be generated is 2 m ;

[0036] Step 2, according to the number of primary inputs n , to generate a set of subscripted odd test codes that meet the maximum total Cartesian distance;

[0037] Step 3. On the basis of obtaining the subscript odd test code set, according to the rule of the maximum Hamming distance, further obtain the subscript even test code set to form an intermediate test set;

[0038] Step 4, intermediate test set and specified test seeds t 0 Perform XOR operation to obtain the final complete test set.

Embodiment 2

[0040] This embodiment is basically the same as Embodiment 1, and its special features are as follows:

[0041] Said step 2 includes:

[0042] Step 2.1, according to the number of primary inputs n , calculate the best distance as the closest n Even number of / 2 σ , to select candidate codes;

[0043] Step 2.2, according to the selected optimal distance σ , set the distance range range = σ ,number of failures fcount =0; random from n Bit binary test code selected from σ The ones place is juxtaposed to 1, and the generated Hamming distance from all 0 yards is σ The test code of the test code is used as the candidate code; if the Hamming distance between the candidate code and all the generated subscript odd test codes is within range Within the range, go to step 2.3; otherwise, fcount Add 1, repeat step 2.2; if fcount exceeds the specified threshold, then relax range scope: range =[ range lower limit -2, range cap+2];

[0044] Step 2.3, use the ca...

Embodiment 3

[0048] In this embodiment, the quasi-complete maximum distance random testing method is based on Microsoft visual studio 2010, and is implemented with C++ object-oriented language as the programming language.

[0049] figure 1 As shown, the complete maximum distance random test method in this paper includes the following steps, and its characteristics are:

[0050] Step 1, read the number of input terminals of the circuit from the text n , the target test set size is 2 m ;

[0051] Step 2, initialize each parameter, generate candidate test codes, and obtain all subscript odd test codes;

[0052] In step 2 mentioned include:

[0053] Step 2.1, use all "0" as the initial test seed t 0 , to calculate the optimal distance even number σ , which has σ The sum of the Cartesian distances between any test code of one "1" and all "0" and all "1" is the largest. when n ≠4 k +2; ( k , n?N )hour, σ =[ n / 2] (even number); when n =4 k +2; ( k , n?N ), σ...

Images