DIAGNOSTIC IMPROVEMENTS FOR MULTI-INSTITUTIONS OF IDENTICAL STRUCTURES

By comparing test results across different structures in a BIST circuit and analyzing mismatches, the method accurately identifies and diagnoses faulty components, enhancing the diagnostic capabilities of BIST circuits.

DE102021128930B4Active Publication Date: 2024-11-14INTERNATIONAL BUSINESS MACHINE CORPORATION
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
DE102021128930
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-12-08
Filing Date
2021-11-08
Publication Date
2024-11-14
Estimated Expiration
2041-11-08

AI Technical Summary

Technical Problem

Existing built-in self-test circuits (BIST) cannot accurately identify which components are defective within a circuit, only indicating if the circuit is defective as a whole.

Method used

The method involves comparing test results from different structures of a BIST circuit, using a control circuit to unload and analyze these results, identifying faulty components by detecting mismatches and logging test loops and shift counters to pinpoint the failing components.

Benefits of technology

Enables precise identification of defective components within a BIST circuit, allowing for targeted diagnosis and fault detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000013_0000
    Figure 00000013_0000
  • Figure 00000014_0000
    Figure 00000014_0000
  • Figure 00000015_0000
    Figure 00000015_0000
Patent Text Reader

Abstract

Procedures, exhibiting: Performing (902) a test on a first structure and a second structure of an in-house self-test circuit, wherein both the first and the second structure have a plurality of flip-flops (105) arranged as a plurality of block chains (104); Unloading (904, 906) a first result of the test from the majority of block chains of the first structure and a second result of the test from the majority of block chains of the second structure; Generating a first signature based on the first result and a second signature based on the second result; Determine that the first structure failed the test based on the fact that the first signature does not match a reference signature, and that the second structure passed the test based on the fact that the second signature matches the reference signature; Comparing the first result with the second result to identify any mismatch between the first and second results; and Determine (908) a faulty flip-flop in the first structure, based on the lack of a match.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUND

[0001] The present invention relates to built-in self-test circuits and in particular to the diagnosis of faulty circuit elements by built-in self-test circuits.

[0002] US 2009 / 0089636A1 describes a method, system, and computer program product for identifying faults in multi-core processors using LBIST (Logic Built-in Self Test) technology.

[0003] US 2005 / 0138501A1 describes a system and a method for testing a first and a second set of electronic devices on a microchip. SUMMARY

[0004] According to one embodiment, a method comprises performing a test on a first structure and a second structure of an embedded self-test circuit. Both the first and second structures contain a plurality of flip-flops arranged as a plurality of block chains. The method also comprises unloading a first test result from the plurality of block chains of the first structure and a second test result from the plurality of block chains of the second structure. The method further comprises determining, based on a discrepancy between the first and second results, that the plurality of block chains of the first structure contains a faulty flip-flop. Further embodiments include a device and a system that execute this method. Brief description of the different views of the drawings Fig. Figure 1 shows an example system; Fig. 2 shows an example of unloading in the system of Fig. 1; Fig. Figure 3 shows an example of unloading in the system of Fig. 1; Fig. 4 shows an example of unloading in the system of Fig. 1; Fig. Figure 5 shows an example of compression in the system of Fig. 1; Fig. Figure 6 shows an example of a distribution circuit in the system of Fig. 5; Fig. Figure 7 shows an example XOR / toggle switch in the system of Fig. 5; Fig. Figure 8 shows an example of unloading in the system of Fig. 1; Fig. 9 is a flowchart of an example procedure in the system of Fig. 1. DETAILED DESCRIPTION

[0005] The invention is described by the features of the independent claims. Embodiments are specified in the dependent claims.

[0006] Built-in self-test circuits (BIST circuits) have structures that allow tests to be performed directly on the circuit logic connected to the BIST circuits. The test results indicate whether the BIST circuits or the circuit logic are faulty. However, these tests cannot identify which components are faulty. This disclosure considers various methods for discharging and analyzing the test results of BIST circuits to determine which components of the BIST circuits or the circuit logic are faulty. In one embodiment, the results of the same test are discharged from different structures of a BIST circuit, and the results are compared. If a discrepancy in the results is detected, a logged test loop and a shift counter indicate the faulty component of the BIST circuit or the circuit logic.the circuit logic that corresponds to the deviation. These methods are used with the . Fig. 1 to 9 are described in more detail.

[0007] One refers to Fig. 1, which shows an exemplary System 100. As in Fig. As shown in Figure 1, the system 100 comprises a control circuit 101, one or more test structures 102, and a circuit 116 containing circuit logic between the flip-flops in the block chains 104. Generally, the control circuit 101 uses the one or more test structures 102 to test not only the block chains 104 but also the circuit 116. The control circuit 101 loads test data into the test structures 102 and then executes the test on the circuit 116. The control circuit 101 then unloads the test results from the test structures 102. The test results can be analyzed to identify fault detection locations in the structures 102. In certain embodiments, the test structures 102 can be logically identical.

[0008] The system 100 can contain any suitable number of control circuits 101, which control any suitable number of test structures 102. In some embodiments, the system 100 contains multiple instances of test structures 102, shown as rows of test structures 102. Each row of test structures is controlled by a control circuit 101. Separate control circuits 101 can be used to control separate rows of test structures 102, or one control circuit 101 can be used to control multiple rows of test structures 102. For clarity, Fig. 1 a control circuit 101 and a series of test structures 102.

[0009] The control circuit 101 contains a phase-locked loop 112, a clock generator 114, and a BIST control engine 116. In general, the phase-locked loop 112 and the clock generator 114 work together to generate one or more clock signals that control other components of the system 100 (e.g., the test structures 102). The engine 116 controls the clocking, the loading of test data, and the unloading of test results from the one or more test structures 102.

[0010] The phase-locked loop 112 generates a signal with a specific frequency. For example, the phase-locked loop 112 can contain a variable-frequency oscillator and a phase detector in a feedback loop. The phase-locked loop 112 uses this feedback loop to generate an output signal with a specific frequency.

[0011] The clock generator 114 can generate one or more clock signals using the output signal of the phase-locked loop 112. For example, the clock generator 114 can include circuit elements that multiply or divide the output signal of the phase-locked loop 112 by one or more factors. In this way, the clock generator 114 generates output signals with frequencies that are proportional to the frequency of the output signal of the phase-locked loop 112. The clock generator 114 transmits these clock signals to other components of the system 100 to control the timing and operation of these components.

[0012] Engine 116 controls the loading of test data and the unloading of test results from test structures 102. As in Fig. As shown in Figure 1, the engine 116 can comprise a processor 118 and a memory 120, which are configured to perform each of the functions or operations of the control circuit 101 described herein. In certain embodiments, the control circuit 101 can determine a fault detection location in the structure 102 by unloading and analyzing the test results from the test structure 102.

[0013] The processor 118 is any electronic circuit, for example, but not limited to, a microprocessor, an application-specific integrated circuit (ASIC), an application-specific instruction set processor (ASIP), and / or a state machine, which is connected to the memory 120 for data transmission and controls the operation of the control circuit 101. The processor 118 can have an 8-bit, 16-bit, 32-bit, 64-bit, or any other suitable architecture. The processor 118 can include an arithmetic logic unit (ALU) for performing arithmetic and logical operations, processor registers that supply the ALU with operands and store the results of the ALU operations, and a control unit that retrieves instructions from memory and executes them by controlling the coordinated operations of the ALU, the registers, and other components.The processor 118 may include other hardware that executes software to control and process information. The processor 118 executes software stored in memory to perform each of the functions described herein. The processor 118 controls the operation and management of the control circuit 101 by processing information (e.g., information received and loaded from the external test system). The processor 118 may be a programmable logic unit, a microcontroller, a microprocessor, any suitable processing unit, or any suitable combination thereof. The processor 118 is not limited to a single processing unit and may include multiple processing units.

[0014] Memory 120 can store data, operating software, or other information for Processor 118, either permanently or temporarily. Memory 120 can comprise a register, built-in array memory, and / or a combination thereof, suitable for storing BIST test sequence formation and data logging information. For example, Memory 120 can comprise random-access memory (RAM), read-only memory (ROM), magnetic storage devices, optical storage devices, or any other suitable information storage device, or a combination thereof. Software represents any suitable set of instructions, logic, or code contained in a computer-readable storage medium. For example, software can be contained in Memory 120, a floppy disk, a CD, or a flash drive.In certain embodiments, the software may include an application that can be executed by the processor 118 to perform one or more of the functions described herein. In certain embodiments, the engine 116 is an integrated state machine for controlling the execution of the test. The engine 116 is set up or loaded by an external test system and does not access external memory during the execution of the test.

[0015] Each test structure 102 comprises a pseudorandom pattern generator (PRPG) 103, one or more block chains 104, each containing one or more flip-flops 105, a channel mask 106, and a multi-input signature register (MISR) 107. For the sake of clarity, only the components of a test structure 102 were shown in Fig. 1 marked, however, the system 100 can include any suitable number of test structures 102 containing these components.

[0016] The control circuit 101 loads test data into one or more PRPGs 103. In some embodiments, the control circuit 101 provides data to the PRPGs 103, and the PRPGs 103 use this data to generate pseudorandom test data. Each PRPG 103 loads the test data into the blockchains 104 of its corresponding test structure 102.

[0017] Each blockchain 104 contains one or more flip-flops 105. The blockchains 104 in a test structure 102 are arranged in parallel. The flip-flops 105 in a blockchain 104 are arranged serially (e.g., chained together in a scan chain). The PRPG 103 of a test structure 102 serially loads test data into the blockchains 104 of the test structure 102. The test data can be loaded simultaneously by all PRPGs 103 into all blockchains 104. Each flip-flop 105 can be used to test a different component or section of the logic circuits between the scan chains in instance 116. For example, while the flip-flops are clocked, the test data is passed from and through the flip-flops 105, stimulating different components of the logic test circuit 116 between the blockchains. Answers are stored in the respective block flip-flops 105 of the test structure 102.In this way, the flip-flops 105 in the block chains 104 can contain the results of the test after the test data has been passed on by the system clocks 114.

[0018] Channel mask 106 and MISR 107 collect and compress the test results from the block chains 104. Channel mask 106 can (via gating) prevent any combination of block chains 104 from reaching MISR 107. MISR 107 reads the results from the unblocked block chains 104 and compresses these results into a signature. Control circuit 101 compares the signature in MISR 107 with a reference signature to determine whether circuit 116 has passed the test. For example, if the signature in MISR 107 matches the reference signature, control circuit 101 can determine that circuit 116 has passed the test. Conversely, if the signature in MISR 107 does not match the reference signature, control circuit 101 can determine that circuit 116 has failed the test.

[0019] As in Fig. As can be seen in Figure 1, certain components of the test structures 102 can be arranged serially in a row. For example, the PRPGs 103 and the MISRs 107 can be serialized so that the control circuit loads data into a first PRPG 103 and MISR 107 (e.g., on the left) and this data is passed on to a last PRPG 103 and MISR 107 (e.g., on the right). In some embodiments, these components do not need to be serialized and instead allow parallel loading and unloading. In some embodiments, the MISRs 107 include a feedback block to enable serial loading and unloading of the MISRs 107.

[0020] This disclosure describes various ways in which the control circuit 101 can unload and analyze the test results from the block chains 104 in order to identify the faulty components in a tested circuit. These methods are described with reference to the Fig. 2 to 8 described.

[0021] Fig. Figure 2 shows an example of unloading in system 100. Fig. 1. As in Fig. As can be seen in Figure 2, the system 100 comprises several instances (shown as rows) of block structures 102. A control circuit 101 controls each instance of the block structures 102. In the example of Fig. Control circuit 101A controls the instance comprising block structures 102A, 102B, and 102C. Control circuit 101B controls the instance comprising block structures 102D, 102E, and 102F. Control circuit 101C controls the instance comprising block structures 102G, 102H, and 102I. System 100 also includes a selector 205, a comparator 206, a selection register 210, a recorder 211, and a database 215. In certain embodiments, control circuits 101A, 101B, and 101C can be replaced by a single control circuit 101 that controls all instances of block structures 102. Alternatively, there can be a separate control circuit for each block structure. In some embodiments, the selection register 210, the recorder 211, and the database 215 are implemented as part of a control circuit 101 and can exchange data with the external test system. For the sake of clarity, certain components in the system 100 are shown in Fig. 2 shown, but not labeled; however, these components share the features shown in Fig. 1. Names given.

[0022] In the example of Fig. 2. Initially, the control circuits 101 execute the same test in each instance of block structures 102. Each block structure 102 can test a different section of the logic circuit between its block chains; however, block structures 102 that are in the same column (e.g., block structures 102A, 102D, and 102G) test the same sections of the logic circuit. Therefore, if the same test is passed through the block structures 102 in a column, it is expected that the MISRs 107 in these block structures 102 will contain the same signatures for a "good" or passed test. After the tests are completed, the test results are compressed into the MISRs 107, forming signatures. These signatures in the MISRs 107 are compared with reference signatures to determine which instances failed the test and which instances passed the test. In the example of Fig. The MISRs 107 indicate that the upper instance failed the test and the middle instance passed. However, these signatures do not necessarily indicate which circuit elements, block chain 104, and flip-flops 105 failed the test, or which test cycle they failed.

[0023] After the passing and failing instances have been identified, the control circuits 101 switch the system 100 into a second phase to diagnose the fault(s) and identify the faulty components. The selection register 210 generates a control signal at the selector 205 to select the outputs of the upper (e.g., failing) and middle (e.g., passing) instances for comparison. Additionally, the control circuits 101 switch the MISRs 107 of the upper and middle instances to operate as shift registers. The control circuits 101 then re-run the test using the upper and middle instances. Generally, the control circuits 101 then unload and compare the test results from the two instances, using the successful instance as a reference. In the example of Fig. 2. The results from the middle instance are compared with the results from the upper instance, since the upper instance failed the initial test and the middle instance passed the initial test, with the results from the middle instance serving as a reference.

[0024] After the test results have been loaded into the blockchains 104, the control circuits 101 generally switch the channel mask 106 so that the test results from a pair of corresponding blockchains 104 in the instances (e.g., the same blockchains 104 in the test structures 102 in the same column) are unloaded. These test results are compared to identify any discrepancies indicating a test failure. The comparison and the results are then logged. The control circuits 101 then switch the channel mask 106 so that the test results from the next pair of corresponding blockchains 104 in the instances are unloaded. These test results are compared to identify any discrepancies indicating a test failure. The comparison and the results are then logged. This process continues until the results of all blockchains 104 in the instances have been unloaded and compared.

[0025] In the example of Fig. The control circuits 101 have configured the channel masks 106 such that the test results from the second block chains 104 are unloaded into the test structures 102B and 102E. The remaining block chains 104 in the instances are blocked. The test results of these block chains 104 are loaded into the respective MISRs 107. Since the MISRs 107 are configured to operate as shift registers, the test results from the block chains 104 are shifted by the MISRs 107 of the instances towards the selector 205, with compression disabled. Based on the control signal generated by the selection register 210, the selector 205 selects the results from the upper and middle instances for comparison by the comparator 206.

[0026] Recorder 211 then records the comparison results as well as information about the test results. If a discrepancy in the test results exists, recorder 211 logs the discrepancy in database 215. Furthermore, recorder 211 increments error counter 214. Recorder 211 also logs the test loop 213 in which the discrepancy occurred. Additionally, based on the position in the sequence of test results where the discrepancy occurred, recorder 211 logs information that identifies the flip-flop 105 of blockchain 104 where the discrepancy occurred. For example, the shift counter 212 can display a value A indicating the position of flip-flop 105 in blockchain 104, a value B indicating blockchain 104 in test structure 102, and a value C indicating test structure 102 in the instance.Using this information, the exact flip-flop 105 where the deviation occurred can be identified.

[0027] In the example of Fig. 2. A discrepancy between a flip-flop 105A and a corresponding flip-flop 105B in the block chains 104 is detected. As a result, the recorder 211 logs the values ​​A, B, and C in the displacement counter 212, which identify the flip-flop 105, the block chain 104, and the test structure 102 where the discrepancy occurred. The recorder 211 also logs the test results in the database 215 that show the discrepancy. The recorder 211 also logs the test loop 213, so that the test data for this test loop can be identified. Using all this information, a control circuit 101 or an administrator can determine the discrepancy in the test results, the circuit elements tested by the flip-flops 105 that triggered the discrepancy, and the test data used to test these faulty circuit elements.As a result, the control circuit 101 and the administrator can locate and diagnose the cause of the fault. In some embodiments, multiple faulty flip-flops 105 in the same block chain 104 or different block chains 104 can be identified by repeating the diagnostic procedure.

[0028] After the results from the pair of blockchains 104 have been unloaded and compared, the control circuits 101 adjust the channel masks 106 to begin unloading the next blockchains 104 in the instances (e.g., the third blockchains 104 in test structures 102B and 102E). The results from the next blockchains 104 are compared and analyzed to identify any deviations. This process can be continued until all deviations in all blockchains 104 in the instances have been unloaded, compared, and analyzed. After the test results of all blockchains 104 have been unloaded and compared, the control circuits 101 and the administrator can check the error counter 214 to determine the number of deviations identified.Furthermore, the control circuits 101 and the administrator can evaluate the displacement counter 212, the test loop 213 and the database 215 to identify the faulty flip-flops 105 and the corresponding faulty circuit elements and to diagnose the cause of the errors.

[0029] Fig. Figure 3 shows an example of unloading in system 100. Fig. 1. In general, this is similar to the in Fig. The 3 methods shown in Fig. The two methods shown differ in that each unloading and comparing in the method of Fig. 3 is carried out on a cut 305 of the flip-flops 105 and not on a block chain 104. As in Fig. As shown in Figure 3, during an unload and comparison, a flip-flop 105 from each blockchain 104 in the upper and middle instances is unloaded and sent to the selector 205. The results of these flip-flops 105 are then compared, and any discrepancies are recorded. The next flip-flops 105 in the blockchains 104 can then be unloaded and compared in a similar manner. This process can continue until all flip-flops 105 in the blockchains 104 have been unloaded and compared. For clarity, certain components in system 100 are shown in Fig. 3 shown, but not labeled; however, these components share the features shown in Fig. 1. Names given.

[0030] As in the example of Fig. 2. The control circuits 101 can perform an initial test on the instances of the test structures 102. The control circuits 101 can then compare the MISRs 107 of the different instances to determine which instances passed the initial test and which instances failed the initial test. In the example of Fig. 3. The control circuits 101 determine that the upper instance failed the initial test and the middle instance passed the initial test. The selection register 210 can then generate a control signal that causes the selector 205 to select the outputs of the upper and middle instances. The control circuits 101 can then toggle the channel mask 106 in the upper and middle instances so that the outputs of all block chains 104 in the test structures 102 of the upper and middle instances are passed. The MISRs 107 can then read a slice 305 of the flip-flops 105 from the block chains 104 of the instances. When the results of these flip-flops 105 have been read into the MISRs 107, the MISRs 107 transfer these results to the selector 205 and the comparator 206. The comparator 206 compares these results, and the recorder 211 records all deviations along with the information belonging to the deviations.The MISRs 107 can then unload the next cut 305 of the flip-flops 105 from the block chains 104 for comparison at the comparator 206. In the example of . Fig. Step 3: A section 305A of flip-flops 105 is unloaded from the upper instance, and a section 305B of flip-flops 105 is unloaded from the middle instance. The flip-flops 105 are transferred to comparator 205 and compared by comparator 206 to detect any discrepancies.

[0031] As in Fig. As shown in Figure 3, the system 100 comprises one or more single-input signature registers 303. When the results are unloaded from the block chains 104, they are compressed into the SISRs 303 to generate a signature for the compared instances. In certain embodiments, the signatures in the SISRs 303 are used to determine which instances passed or failed the initial test, instead of initially executing the test and comparing the MISRs 107 with the reference signature. For example, the test can be executed on the instances, and then the test results are unloaded from the cuts of the flip-flops 105 and compared. The test results are also compressed into the SISRs 303 to generate a signature for the instances.At the end of the discharge comparison, the signatures in the SISRs 303 can be compared with the reference signature to determine which instances were successful and which failed. The control circuits 101 can then determine for each deviation which instance contained the faulty flip-flop 105 and which instance contained the successful flip-flop 105.

[0032] Fig. Figure 4 shows an example of unloading in system 100. Fig. 1. In general, this is similar to the in Fig. The 4 methods shown in Fig. The two methods shown, with the difference that the method in Fig. 4 the flip-flops 105 of the block chains 104 discharge serially. As in Fig. As shown in Figure 4, the blockchains 104 in a test structure 102 can be switched so that test results are unloaded to the channel mask 106 or serially through the blockchains 104. For example, test structure 102A comprises blockchains 104A, 104B, and 104C. Blockchains 104A, 104B, and 104C can unload their test results in parallel through the channel mask 106 into the MISR 107, for example, during the initial test execution, so that the signature in the MISR 107 can be compared with a reference signature to determine whether test structure 102A has passed the test. Additionally, blockchains 104A, 104B, and 104C can be switched so that they serially unload their results over each other. For example, blockchain 104A can unload its results into blockchain 104B.Blockchain 104B can unload its test results into blockchain 104C, and blockchain 104C can unload its test results into MISR 107 via channel mask 106. Serial unloading can be used when unloading and comparing test results from test structures 102. For clarity, certain components in system 100 are shown in [reference missing]. Fig. 4 shown, but not labeled; however, these components share the features shown in Fig. 1. Names given.

[0033] During test execution, the control circuits 101 can send test data from the PRPG 103 to the block chains 104. The test can then be executed, and the results can be stored in the block chains 104. During unloading and comparison, the control circuits 101 can switch the multiplexers 404 so that data is unloaded serially through the block chains 104. Additionally, the control circuits 101 can switch the channel mask 106 so that information is passed from the last block chain 104 to the subsequent test structure 102. The control circuit 101 can also switch the MISR 107 to operate as a shift register, so that the result from the last block chain 104 is shifted to other test structures 102 in the direction of the selector 205.The control circuits 101 also switch a multiplexer 402 on each test structure 102 such that the test results are moved from a previous MISR 107 to the block chains 104 of the test structure 102. In this way, the test results are serially unloaded from the block chains 104 through the block chains 104 of all test structures 102. For example, the test results of a test structure 102A are moved by the MISR 107 in test structure 102A towards a test structure 102B. The multiplexer 402 in test structure 102B then sends these test results to the block chains 104 of test structure 102B. The block chains 104 in test structure 102B then serially unload these test results via the channel mask 106 and the MISR 107 of test structure 102B. MISR 107 in test structure 102B moves the test results to selector 205. In this way, the test results are moved serially from the instance to selector 205.Similarly, the test results from another instance containing test structures 102C and 102D are also moved to selector 205. Comparator 206 compares these test results, and recorder 211 identifies and logs deviations. Additionally, the results can be compressed in SISRs 303, as in the example of... Fig. As described in section 2, the signatures in the SISRs 303 can be used to determine which instance contained faulty flip-flops 105 or faulty components, instead of running the initial test and comparing the signatures in the MISRs 107 with the reference signature. As described in Fig. As shown in Figure 4, each instance can contain any suitable number of test structures 102. For clarity, each test structure 102 has been shown with three blockchains 104, but each test structure 102 can contain any suitable number of blockchains 104.

[0034] The Fig. Figures 5 to 7 show an exemplary improvement in compression in the system 100 of Fig. 1, which can be performed on one or more of the unloading procedures described here. In general, compression during unloading is performed by distributing the results of several blockchains 104 in a test structure 102 and then rotating and combining the distributed results with the distributed results of the other test structures 102 in an instance. The compressed results for the instances are then compared to identify discrepancies. Recorder 511 may include enhancements that allow Recorder 511 to decompress or interpret the comparison result. For clarity, certain components in System 100 are in Fig. 5 shown, but not labeled; however, these components share the features shown in Fig. 1. Names given.

[0035] In the example of Fig. In section 5, test structure 102 comprises a distribution circuit 501 and an XOR / toggle circuit 502. Generally, the distribution circuit 501 distributes the test results from the block chains 104 within test structure 102. The XOR / toggle circuit 502 then combines the result from the distribution circuit 501 with the toggle results from the previous test structures 102 within the instance. The XOR / toggle circuit 502 then passes the combined result to the MISR 107. The MISR 107 passes the combined result via a multiplexer 503 to the next test structure 102 within the instance, where the combined result is again combined with the toggle distributed result from that test structure 102. The combined result of the instance is finally passed to the selector 205. The comparator 206 then compares the combined results from the two instances to identify any discrepancies.The combined results from the instances can be compressed in the MISRs 508 for subsequent analysis to determine which instance contains the faulty flip-flop 105 or the faulty circuit element.

[0036] Recorder 511 records the combined results from the instances and logs information about any detected deviations. Recorder 511 may include enhancements that allow it to decompress or interpret the combined results from the instances. For example, Recorder 511 can decompress or interpret the combined results to identify the test structure 102, the blockchain 104, and the faulty flip-flop 105. The distribution circuit 501 and the XOR / toggle circuit 502 are used in relation to the Fig. 6 and Fig. 7 described in more detail. In some embodiments, after a test, the combined results can be decompressed or interpreted by an external test system to identify faulty flip-flops 105.

[0037] Fig. Figure 6 shows an example distribution circuit 501 in the System 100 of Fig. 5. As in Fig. As shown in Figure 6, the distribution circuit 501 contains several XOR gates that receive inputs from one or more block chains 104 of the test structure 102. The results from the block chains 104 are distributed to the various XOR gates in the distribution circuit 501. As shown in Figure 6, the distribution circuit 501 contains several XOR gates that receive inputs from one or more block chains 104 of the test structure 102. The results from the block chains 104 are distributed to the various XOR gates in the distribution circuit 501. Fig. As shown in Figure 6, the outputs of the block chains 104 (e.g., outputs 0 to 31) are sequentially provided as the first inputs to the XOR gates of the distribution circuit 501. Each XOR gate also receives inputs from two other block chains 104. The two additional block chains 104 for any given XOR gate can be organized such that each XOR gate receives a sequence of outputs from a different combination of block chains 104. For example, a first XOR gate (XOR 0) receives inputs from block chains 0, 1, and 2, while a second XOR gate (XOR 1) receives inputs from block chains 1, 3, and 31. The outputs of the XOR gates are then output by the distribution circuit 501.

[0038] Furthermore, the 501 distribution circuit includes an enable signal connected to the second and third inputs of each XOR gate. When the enable signal is low, each XOR gate passes its first input through, meaning the XOR gates pass the outputs of the 104 blocks sequentially. When the enable signal is high, the 501 distribution circuit begins distributing the outputs of the 104 blocks by allowing the XOR gates to receive inputs from their second and third inputs. In other words, when the enable signal is low, the 501 distribution circuit does not distribute the outputs of the 104 blocks. When the enable signal is high, the 501 distribution circuit does distribute the outputs of the 104 blocks.

[0039] Fig. Figure 7 shows an example of a 502 XOR / toggle switch in System 100. Fig. 5. As in Fig. As shown in Figure 7, the XOR / toggle circuit 502 contains a series of XOR gates that receive the outputs of the distribution circuit 501 of a test structure 102. These XOR gates also receive the outputs of a previous test structure 102. The output of the previous test structure 102 may be toggled by 1 bit (e.g., XOR 0 receives the last bit (p31) of the results from the previous test structure 102, and XOR 1 receives the first bit (p0) of the results from the previous test structure 102). The output of the XOR gates is then sent to the next test structure 102. In certain embodiments, the results of an instance are compressed before comparison by distributing the results from the block chains 104 using the distribution circuits 501 and then combining them with the toggled results from previous test structures 102 using the XOR / toggle circuits 502.

[0040] Fig. Figure 8 shows an example of unloading in the system 100 of Fig. 1 such that all blockchains for each instance can be chained together into a serial scan chain during the diagnostic unload procedure. In general, the in Fig. The eight test structures shown in Figure 802 can be used with both BIST structures and deterministic test data. For example, deterministic test data can be loaded serially and simultaneously into all instance channels of the test structures 802. A feedback loop 803 can recirculate the outputs through the test structures 802 for compression purposes. The results from the instances can be selected and compared by the comparator 507. The recorder 811 can then log any discrepancies in the test results.

[0041] Fig. Figure 9 is a flowchart of an exemplary procedure 900 in system 100 of the Fig. 1. In general, one or more control circuits 101 can perform the steps of procedure 900. In certain embodiments, the control circuits 101 discharge and analyze test results by performing procedure 900 to identify faulty flip-flops 105 and faulty circuit components. Furthermore, the control circuits 101 diagnose the cause of a faulty flip-flop 105 and faulty circuit element.

[0042] In step 902, the control circuit 101 performs a test on a first and a second test structure 102 of an integrated self-test circuit. Executing the test can involve sending deterministic or pseudorandom test data through the test structures 102. The test data can be sent through flip-flops 105, which are organized as block chains 104. Circuit elements can be tested using the test data in a flip-flop 105, and the results of this test are then loaded back into the flip-flop 105. The flip-flop 105 passes the test data to another flip-flop 105 before the test results are loaded into the flip-flop 105. The test is complete when the test data has been passed through the flip-flops 105 and the flip-flops 105 contain the test results.

[0043] In step 904, the control circuit 101 unloads a first result from a first test structure 102. Simultaneously, in step 906, the control circuit 101 unloads a second result from a second test structure 102. The control circuit 101 can unload these test results using any suitable method. For example, the control circuit 101 can distribute these test results across a blockchain 104 via a blockchain according to the method described in Fig. The two methods shown are discharged. The control circuit 101 can output the test results as cuts 305 of the flip-flops 105 according to the method shown in Fig. The 3 methods shown are discharged. The control circuit 101 can also output the test results serially according to the procedure shown. Fig. The four methods shown are discharged. After the test results have been discharged, they are compared to identify any deviations.

[0044] In step 908, the control circuit 101 determines, based on a discrepancy between the first and second results, that the first structure 102 contains a faulty flip-flop 105. For example, the control circuit 101 may implement a recorder 211 that records deviations and information regarding these deviations in a displacement counter 212. This allows the control circuit 101 to identify the test structure 102, the blockchain 104, and the flip-flop 105 associated with the deviation. After identifying the faulty flip-flop 105, the control circuit 101 can determine the circuit element that was tested using the test data in the faulty flip-flop 105. In this way, the control circuit 101 can identify the faulty circuit element. Furthermore, the recorder 211 may have logged the test loop 213 in which the deviation occurred.The control circuit 101 can use this information to identify the test data that caused the deviation. The recorder 211 can also log the results of the failed tests. The control circuit 101 can examine the results of the failed tests to diagnose a cause of the deviation. In this way, in certain embodiments, the control circuit 101 can identify a faulty flip-flop 105 and a faulty circuit element and diagnose a cause of the fault.

[0045] The description of the various embodiments of the present invention is given for illustrative purposes only and is not to be considered exhaustive or limited to the disclosed embodiments. Many modifications and variants are obvious to those skilled in the art without deviating from the scope and spirit of the described embodiments. The terminology used here has been chosen to best describe the principles of the embodiments, their practical application or technical improvement over technologies available on the market, or to enable other skilled persons to understand the embodiments disclosed herein.

[0046] The foregoing referred to embodiments presented in this disclosure. However, the scope of this disclosure is not limited to specific described embodiments. Instead, any combination of features and elements is considered, regardless of whether they relate to other embodiments or not, for the implementation and practice of the embodiments under consideration. Although embodiments disclosed herein may offer advantages over other possible solutions or over the prior art, the extent to which a particular advantage is achieved by a given embodiment does not limit the scope of this disclosure.Therefore, the aspects, features, embodiments, and advantages described herein are merely examples and are not considered elements or limitations of the appended claims unless expressly stated in one or more claims. Likewise, reference to "the invention" should not be construed as a generalization of any inventive subject matter disclosed herein and should not be regarded as an element or limitation of the appended claims unless expressly stated in one or more claims.

[0047] Aspects of the present invention may take the form of an embodiment entirely in hardware, an embodiment entirely in software (for example, firmware, local software, microcode, etc.), or an embodiment that combines software and hardware aspects, all of which may be generally referred to herein as a “circuit”, “module”, or “system”.

[0048] The present invention may be a system, a method, and / or a computer program product at any possible level of technical integration. The computer program product may comprise a computer-readable storage medium (or media) containing computer-readable program instructions to induce a processor to execute aspects of the present invention.

[0049] A computer-readable storage medium can be a physical unit capable of holding and storing instructions for use by a system for executing instructions. For example, a computer-readable storage medium can be an electronic storage unit, a magnetic storage unit, an optical storage unit, an electromagnetic storage unit, a semiconductor storage unit, or any suitable combination thereof, without limitation. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), and erasable programmable read-only memory (EPROM).Flash memory), static random-access memory (SRAM), portable compact storage disk-read-only memory (CD-ROM), DVD (digital versatile disc), USB flash drive, floppy disk, a mechanically coded unit such as punched cards or raised structures in a groove on which instructions are stored, and any suitable combination thereof. A computer-readable storage medium shall not, in its use herein, be understood as volatile signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., light pulses traveling through an optical fiber cable), or electrical signals transmitted by a wire.

[0050] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to individual data processing units or, via a network such as the internet, a local area network, a wide area network, and / or a wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission lines, wireless transmission, routing computers, firewalls, switching units, gateway computers, and / or edge servers. A network adapter card or network interface in each data processing unit receives computer-readable program instructions from the network and forwards them for storage on a computer-readable storage medium within the respective data processing unit.

[0051] Computer-readable program instructions for executing work steps of the present invention may be assembler instructions, ISA (Instruction Set Architecture) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuits, or either source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., as well as procedural programming languages ​​such as the programming language "C" or similar programming languages.The computer-readable program instructions can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In the latter case, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be established with an external computer (for example, via the internet using an internet service provider).In some embodiments, electronic circuits, including, for example, programmable logic circuits, field programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), can execute computer-readable program instructions by using state information from the computer-readable program instructions to personalize the electronic circuits to implement aspects of the present invention.

[0052] Aspects of the present invention are described herein with reference to flowcharts and / or block diagrams or charts of methods, devices (systems), and computer program products according to embodiments of the invention. It is pointed out that each block of the flowcharts and / or block diagrams or charts, as well as combinations of blocks in the flowcharts and / or block diagrams or charts, can be executed by means of computer-readable program instructions.

[0053] These computer-readable program instructions can be provided to a processor of a computer or other programmable data processing device to create a machine, such that the instructions executed via the processor of the computer or other programmable data processing device generate means for implementing the functions / steps specified in the block(s) of the flowcharts and / or block diagrams or charts.These computer-readable program instructions may also be stored on a computer-readable storage medium capable of controlling a computer, programmable data processing device, and / or other units to function in a particular manner, such that the computer-readable storage medium on which instructions are stored has a manufactured product, including instructions that implement aspects of the function / step specified in the block(s) of the flowchart and / or block diagrams or charts.

[0054] The computer-readable program instructions can also be loaded onto a computer, other programmable data processing device, or other unit to cause the execution of a series of process steps on the computer or other programmable device or other unit in order to generate a process executed on a computer, such that the instructions executed on the computer, other programmable device, or other unit implement the functions / steps specified in the block(s) of the flowcharts and / or block diagrams or charts.

[0055] The flowcharts and block diagrams or charts in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this context, each block in the flowcharts or block diagrams or charts can represent a module, segment, or part of instructions that includes one or more executable instructions for performing the specific logical function(s). In some alternative embodiments, the functions specified in the blocks may occur in a different sequence than shown in the figures.For example, two blocks shown consecutively may in reality be executed essentially as one step, simultaneously, substantially simultaneously, or in a partially or completely overlapping manner, or the blocks may sometimes be executed in reverse order depending on the relevant functionality. It should also be noted that each block in the block diagrams or flowcharts, as well as combinations of blocks in the block diagrams or flowcharts, can be implemented by special hardware-based systems that perform the specified functions or steps, or by combinations of special hardware and computer instructions.

[0056] While the foregoing relates to embodiments of the present invention, other and further embodiments of the invention may be conceived without deviating from its basic scope, and its scope is defined by the following claims.

Claims

[1] Method comprising: Performing (902) a test on a first structure and a second structure of a built-in self-test circuit, wherein both the first and second structures comprise a plurality of flip-flops (105) arranged as a plurality of block chains (104); Unloading (904, 906) a first result of the test from the plurality of blockchains of the first structure and a second result of the test from the plurality of blockchains of the second structure; generating a first signature based on the first result and a second signature based on the second result; Determining that the first structure failed the test on the basis that the first signature does not match a reference signature, and that the second structure passed the test on the basis that the second signature matches the reference signature; Comparing the first result with the second result to identify a mismatch between the first result and the second result; and Determining (908) a faulty flip-flop in the first structure based on the mismatch. [2] The method of claim 1, further comprising switching multi-input signature registers (107), MISRs, of the first and second structures to operate as shift registers. [3] The method of claim 1, wherein unloading the first and second results comprises: Unloading a first blockchain of the first structure and a first blockchain of the second structure; and Unloading a second blockchain of the first structure and a second blockchain of the second structure after unloading the first blockchain of the first structure and the first blockchain of the second structure. [4] The method of claim 1, wherein unloading the first and second results comprises: Unloading a first flip-flop from each block chain of the plurality of block chains of the first and second structures; and Unloading a second flip-flop from each blockchain of the plurality of blockchains of the first and second structures after unloading the first flip-flop from each blockchain of the plurality of blockchains of the first and second structures. [5] The method of claim 1, wherein unloading the first and second results comprises: serially unloading each blockchain of the plurality of blockchains of the first structure; and serially unloading each blockchain of the plurality of blockchains of the second structure. [6] The method of claim 1, wherein unloading the first result comprises unloading each flip-flop of a first blockchain of the plurality of blockchains of the first block such that an output of each flip-flop is input to an XOR gate with outputs of at least two other flip-flops of the first blockchain of the plurality of blockchains of the first block. [7] The method of claim 1, further comprising loading deterministic test data serially and simultaneously into the first and second structures. [8] Device comprising: a memory (120); and a hardware processor (118) connected to the memory for data transfer, the hardware processor (118) being configured to: Performing a test on a first structure and a second structure of a built-in self-test circuit, wherein each of the first and second structures comprises a plurality of flip-flops arranged as a plurality of block chains; Unloading a first result of the test from the plurality of blockchains of the first structure and a second result of the test from the plurality of blockchains of the second structure; generating a first signature based on the first result and a second signature based on the second result; Determining that the first structure failed the test on the basis that the first signature does not match a reference signature, and that the second structure passed the test on the basis that the second signature matches the reference signature; Comparing the first result with the second result to identify a mismatch between the first result and the second result; and Determine a faulty flip-flop in the first structure based on the mismatch. [9] The apparatus of claim 8, wherein the hardware processor (118) is further configured to switch multi-input signature registers, MISRs, of the first and second structures to operate as shift registers. [10] The apparatus of claim 8, wherein unloading the first and second results comprises: Unloading a first blockchain of the first structure and a first blockchain of the second structure; and Unloading a second blockchain of the first structure and a second blockchain of the second structure after unloading the first blockchain of the first structure and the first blockchain of the second structure. [11] The apparatus of claim 8, wherein unloading the first and second results comprises: Unloading a first flip-flop from each block chain of the plurality of block chains of the first and second structures; and Unloading a second flip-flop from each blockchain of the plurality of blockchains of the first and second structures after unloading the first flip-flop from each blockchain of the plurality of blockchains of the first and second structures. [12] The apparatus of claim 8, wherein unloading the first and second results comprises: serially unloading each blockchain of the plurality of blockchains of the first structure; and serially unloading each blockchain of the plurality of blockchains of the second structure. [13] The apparatus of claim 8, wherein unloading the first result comprises unloading each flip-flop of a first blockchain of the plurality of blockchains of the first block such that an output of each flip-flop is input to an XOR gate with outputs of at least two other flip-flops of the first blockchain of the plurality of blockchains of the first block. [14] System comprising: a built-in self-test circuit; and a test engine having a hardware processor configured to: Performing a test on a first structure and a second structure of the built-in self-test circuit, wherein each of the first and second structures comprises a plurality of flip-flops arranged as a plurality of block chains; Switching multi-input signature registers, MISRs, of the first and second structures to operate as shift registers; Unloading a first result of the test from the plurality of blockchains of the first structure and a second result of the test from the plurality of blockchains of the second structure; generating a first signature based on the first result and a second signature based on the second result; Determining that the first structure failed the test on the basis that the first signature does not match a reference signature, and that the second structure passed the test on the basis that the second signature matches the reference signature; Comparing the first result with the second result to identify a mismatch between the first result and the second result; and Determine a faulty flip-flop in the first structure based on the mismatch. [15] The system of claim 14, wherein unloading the first and second results comprises: Unloading a first blockchain of the first structure and a first blockchain of the second structure; and Unloading a second blockchain of the first structure and a second blockchain of the second structure after unloading the first blockchain of the first structure and the first blockchain of the second structure. [16] The system of claim 14, wherein unloading the first and second results comprises: Unloading a first flip-flop from each block chain of the plurality of block chains of the first and second structures; and Unloading a second flip-flop from each blockchain of the plurality of blockchains of the first and second structures after unloading the first flip-flop from each blockchain of the plurality of blockchains of the first and second structures. [17] The system of claim 14, wherein unloading the first and second results comprises: serially unloading each blockchain of the plurality of blockchains of the first structure; and serially unloading each blockchain of the plurality of blockchains of the second structure. [18] The system of claim 14, wherein unloading the first result comprises unloading each flip-flop of a first blockchain of the plurality of blockchains of the first block such that an output of each flip-flop is input to an XOR gate with outputs of at least two other flip-flops of the first blockchain of the plurality of blockchains of the first block.

Citation Information

Patent Citations

  • System and method for testing electronic devices on a microchip

    US20050138501A1

  • Method and Apparatus for Logic Built In Self Test (LBIST) Fault Detection in Multi-Core Processors

    US20090089636A1