JP2014092832A - Flash memory deterioration inspection device, deterioration inspection method, and deterioration inspection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently patrol the entire flash memory without waste.SOLUTION: A flash memory deterioration inspection device comprises: an HW timer 22 that is timer means from which time information can be obtained; determination means 31 that obtains the time information from the HW timer 22 and determines a page for which error check is performed, by operation processing on the basis of the obtained time information; and processing means 32 that specifies the page which is determined by the determination means 31, reads stored data, and performs error check.

Description

  The present invention relates to a flash memory deterioration inspection apparatus, a deterioration inspection method, and a deterioration inspection program.

  The following characteristics are known for the NAND flash memory. The NAND flash memory has a structure in which a plurality of blocks are included and each block includes a plurality of pages. Erasing is performed in units of blocks, and the erasing operation is performed by applying a voltage to the P-type semiconductor layer and extracting electrons from the floating gate.

  On the other hand, information is written by injecting electrons into the floating gate by the quantum tunnel effect. In this case, the N-type semiconductor that is the circuit board is set to the ground potential, and a write voltage is applied to the control gate by a minute current. Electrons stored in the floating gate store information. Here, writing is performed on a page-by-page basis, and a write operation is performed by simultaneously applying a write voltage to the control gates for all cells in the same page.

  Further, the read operation is performed in units of pages. A voltage is applied to the control gates of unselected cells to make them conductive. When there is a certain amount of electrons in the floating gate of the selected cell, current does not flow so much between the source and drain, and this state is set to “0”. Further, when there is no fixed amount of electrons in the floating gate, a current flows relatively between the source and the drain, and this state is set to “1”.

  As described above, the erasing operation can be performed only in units of blocks including a plurality of pages. Further, since overwriting cannot be performed in one operation, it is necessary to perform writing after erasing. In the case of -Level-Cell), it is necessary to once read the contents of all 64 pages of one block outside the NAND flash memory and temporarily store them, and erase all 64 pages of one block. After the necessary rewrite processing is performed in the external storage area of the NAND flash memory, an operation of rewriting the erased block is performed.

  Further, the following items are known for the lifetime of the NAND flash memory. First, in the NAND flash memory, there is an upper limit on the number of rewritable times, and the retention period of recorded contents is finite (maximum 10 to several decades), and information written due to deterioration is lost sometime. In addition, the NAND type is more likely to deteriorate than the NOR type because of the circuit structure. Furthermore, it is said that when used for applications in which data is always recorded, abrupt deterioration (bad blocks) occurs due to characteristics, and the product life is expected to be significantly shortened.

  In general, when the gate oxide film deteriorates by reading and writing to the memory cell, the charge accumulation amount deviates from the original design value, and the difference between “0” and “1” cannot be determined. It is also known that deterioration called “Read Disturb” proceeds because a voltage is applied with the operation of a neighboring cell even in a block in which reading and writing are not performed at all.

  Here, as an example, for example, when the number of data erasures is 0, the number of read times that data destruction is likely to start is about 100,000 times. For example, when the number of data erasures is 60,000 times, The number of readings with a high possibility of starting data destruction is about 10,000 times. When this relationship is graphed, a characteristic line S in FIG. 5 is obtained.

  As is apparent from the above, in the NAND flash memory, when data rewrite and erase are repeated, the cell deteriorates and data cannot be written. For this reason, if data writing / erasing is concentrated only on a specific block, the life of that block will be reached earlier, but it is also considered to use a flash memory avoiding one block that has reached the end of its life. In many cases, the flash memory has a lifetime of 1 block.

  In order to avoid this phenomenon, a technique called wear leveling is used. There are several methods for wear leveling, but in a storage medium using NAND flash memory, the address signal from the outside of the memory chip is converted into a different address inside the chip, and the number of write / erase times in each block is equalized. The technique of making it become is widely used.

  In order to prevent the read disturb error and perform the bad block processing efficiently, a refresh management table is provided in the flash memory device, and the data read count and erase count for each physical block are stored and updated as needed. When the number of data read times reaches a predetermined threshold value, the flag “Low” is set in the physical block. Further, a flag “High” is set for a physical block in which an ECC error is found during data reading. A storage device is known in which a refresh management table is searched as appropriate, a block with a flag “Low” is refreshed, and a block with a flag “High” is subjected to bad block processing (see Patent Document 1).

JP 2009-2224013 A

  However, in the storage device described above, the process is centered on finding a bad block, and the number of write / erase times of each block does not necessarily equalize. In addition, a process called patrol is performed to read the flash memory page periodically corresponding to the deterioration and check the deterioration state of the data, but it takes a lot of time to check the entire area, It is difficult to confirm the entire area within the time when patrol is allowed.

  The present invention has been made in view of the current situation regarding such a flash memory, and an object of the present invention is to provide a flash memory capable of efficiently patroling the entire flash memory without waste and improving data reliability. It is to provide a deterioration inspection apparatus, a deterioration inspection method, and a deterioration inspection program.

  The flash memory deterioration inspection apparatus according to the present invention is a time counting unit capable of extracting time information, and a time determination unit that extracts time information from the time measuring unit and determines a page on which error checking is performed based on the obtained time information. And a processing unit for designating a page determined by the determination unit and reading stored data and performing an error check.

  In the degradation inspection apparatus for flash memory according to the present invention, the processing means reads the stored data by sequentially designating the page of the offset value in each block, using the page determined by the determining means as an offset value. It is characterized by performing.

  The degradation inspection apparatus for flash memory according to the present invention comprises counting means for determining the number of error bits when performing error checking by the processing means, and the processing means compares the number of error bits counted by the counting means with a first threshold value. However, when the number of error bits is larger than the first threshold, the block related to the error check is refreshed.

  In the degradation inspection apparatus for a flash memory according to the present invention, the processing means, when the number of error bits is equal to or less than the first threshold value, the number of error bits counted by the counting means and a second threshold value smaller than the first threshold value. If the number of error bits is larger than the second threshold, an error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check. It is characterized by.

  The flash memory deterioration inspection method according to the present invention refers to a time measuring means capable of extracting time information, a time extracting step for extracting time information from the time measuring means, and an error check by arithmetic processing based on the obtained time information. A determination step for determining a page to be performed, and a processing step for specifying the page determined in the determination step and reading stored data and performing an error check.

  In the flash memory degradation inspection method according to the present invention, in the processing step, the page determined in the determination step is used as an offset value, and the stored data is read out by sequentially specifying the page of the offset value in each block. It is characterized by performing.

  The degradation inspection method for a flash memory according to the present invention includes a counting step for obtaining the number of error bits at the time of an error check in a processing step, and the processing step compares the number of error bits counted in the counting step with a first threshold value. However, when the number of error bits is larger than the first threshold, the block related to the error check is refreshed.

  According to the flash memory degradation inspection method of the present invention, in the processing step, when the number of error bits is equal to or less than the first threshold, the number of error bits counted by the counting means and the second threshold smaller than the first threshold. If the number of error bits is larger than the second threshold, an error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check. It is characterized by.

  A program for inspecting deterioration of a flash memory according to the present invention is a device that designates a page for a flash memory in which data is stored, reads the stored data, performs error checking, and inspects the deterioration of the flash memory. The computer used is determined by referring to a clock from which time information can be extracted, time information is extracted, and a page determined by the determining means is determined based on the obtained time information, and a page on which error checking is performed by arithmetic processing. Is designated, and the stored data is read and error checking is performed, and the processing unit functions.

  In the program for inspecting deterioration of a flash memory according to the present invention, the processing means uses the page determined by the determining means as an offset value and reads out the stored data by sequentially specifying the page of the offset value in each block. It is characterized by performing a check.

  The program for inspecting deterioration of a flash memory according to the present invention further causes the computer to function as counting means for obtaining the number of error bits when performing error checking by the processing means, and the processing means includes the number of error bits counted by the counting means. The first threshold value is compared, and if the number of error bits is larger than the first threshold value, the block related to the error check is refreshed.

  In the program for inspecting deterioration of a flash memory according to the present invention, when the number of error bits is equal to or less than the first threshold value, the processing means has a second error bit number counted by the counting means and a second value smaller than the first threshold value. When the threshold value is compared and the number of error bits is larger than the second threshold value, an error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check. It is characterized by that.

  In the present invention, the time information is taken out from the time measuring means, the page to be checked for error is determined by arithmetic processing based on the obtained time information, the determined page is designated, the stored data is read, and the error check is performed. As a result, the probability that a random page is designated increases, and the entire flash memory can be patroled efficiently and efficiently, and the reliability of data can be improved.

  In the present invention, since the determined page is used as an offset value and the stored data is read by sequentially designating the page of the offset value in each block, the error check is performed. The patrol can proceed so as to read the data of the same page in the adjacent block, and the entire flash memory can be patroled efficiently without waste.

  In the present invention, the number of error bits is obtained at the time of error checking, the counted number of error bits is compared with the first threshold value, and refresh processing is performed based on the magnitude of the error bit number with respect to the first threshold value. Error checking is performed when necessary without checking the data of all the pages, and the reliability of the data can be improved.

  In the present invention, the counted number of error bits is compared with a second threshold value that is smaller than the first threshold value, and when the error bit number is larger than the second threshold value, the erase count and reading in the block related to the error check are read. The error check is performed on all pages in the block related to the error check based on the number of times, so that no error has occurred until all the pages in the block are refreshed, but all pages need to be inspected. It is possible to improve the reliability of data by performing error checking on all pages in a detailed manner.

1 is a block diagram of a main part of a computer system employing a flash memory deterioration inspection apparatus according to an embodiment of the present invention. The figure which shows the procedure of the page reading at the time of the error check by the degradation inspection apparatus of the flash memory which concerns on embodiment of this invention. 6 is a flowchart showing a page read operation during patrol by the flash memory deterioration inspection apparatus according to the embodiment of the present invention. 6 is a flowchart showing a page read operation during patrol by the flash memory deterioration inspection apparatus according to the embodiment of the present invention. The figure which shows the graph of the deterioration characteristic by reading and erase | elimination of flash memory.

  Embodiments of a flash memory deterioration inspection apparatus, a deterioration inspection method, and a deterioration inspection program according to the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a part of a computer system that employs a flash memory deterioration inspection apparatus according to an embodiment. This system is composed of software 1 and hardware 2. The software 1 is stored in a main memory or an external storage device in which a program is stored, and is called by the main memory and executed by the processor.

  The software 1 includes data access software (application program) 11 that calls a driver to request writing or reading to a flash memory, and flash memory driver software (hereinafter referred to as a driver) 12. The driver 12 is software for controlling the flash memory. In addition to writing or reading data in response to a request from the application program, the driver 12 constitutes an embodiment of a program for inspecting deterioration of the flash memory according to the present invention. It has a function to improve the reliability.

  The hardware 2 includes a processor 21 that operates the driver 12 and performs overall control of each unit. The processor 21 includes an HW (hardware) timer 22 which is a time measuring means capable of extracting time information, a volatile memory 23 used when the driver 12 operates, and a flash memory controller 24 for accessing the flash memory 25. Is connected. Further, a flash memory 25 that is a control target is connected to the flash memory controller 24.

  The driver 12 includes a determination unit 31, a processing unit 32, and a counting unit 33. The determining unit 31 extracts time information from a timer that is a time measuring unit, and determines a page on which an error check is performed by arithmetic processing based on the obtained time information. The processing means 32 designates the page determined by the determination means 31, reads the stored data, and performs an error check. Preferably, the processing unit 32 reads the stored data by sequentially designating the page of the offset value in each block using the page determined by the determining unit 31 as an offset value, and performs an error check.

  The counting means 33 can be configured appropriately. The counting unit 33 obtains the number of error bits when the processing unit 32 performs an error check. In this case, the processing means 32 compares the number of error bits counted by the counting means 33 with the first threshold A, and when the number of error bits is larger than the first threshold A, the block for the error check is processed. Perform a refresh. Here, ECC (Error Check & Correction) can be adopted for the error check.

  More preferably, when the number of error bits is equal to or less than the first threshold A, the processing unit 32 compares the number of error bits counted by the counting unit 33 with a second threshold B that is smaller than the first threshold. If the number of error bits is larger than the second threshold value B, an error check is performed on all pages in the block related to the error check based on the erase count and read count in the block related to the error check. The number of times of erasure and the number of times of reading in each block (reading all the pages in the block to be once) can be obtained by the driver 12 and stored in the volatile memory 23.

  For example, as shown in FIG. 2, the flash memory 25 is configured by a set of a plurality of blocks each including a plurality of pages 1 to n. Here, although three blocks are shown, the entire capacity of the flash memory 25 is several Gbytes, and each block includes tens to hundreds of pages. In the computer system configured as described above, the driver 12 operates as each unit by executing a program corresponding to the flowcharts shown in FIGS. 3 and 4, and the operation will be described based on this flowchart. To do.

When the flash memory deterioration inspection program (deterioration inspection system) is started when necessary, the flowchart of FIG. 3 starts, and time information is acquired from the HW timer 22 (S11). Next, the address (physical address) of the error check page is calculated and processed based on the acquired time information (S12). As the arithmetic processing, the time information (bit string) held by the HW timer 22 is set to X, and the value of the information X is set to a power value (m) so that the value is sufficiently larger than the total number of pages in the flash memory. In the power of times, X ^m ) is generated, and the power is divided by the total number of pages to obtain a remainder, and this remainder is used as the physical address of the error check page.

As another method of arithmetic processing, all the “values held by the HW timer 22” at each access to the flash memory a certain number of times after the system is started are multiplied to determine the flash memory used. It is also possible to generate a sufficiently large value with respect to the total number of pages and then obtain the remainder of the division of the value and the total number of physical pages.
Furthermore, as a third method, the management information that the flash memory control software stores in the flash memory irregularly includes the hardware timer value at the time of saving, and always holds one or more timer values in the management information. .
At system startup, multiply the "value held by the hardware timer" by the "timer value held in the management information" to give a value that is large enough for the total number of pages in the flash memory being used. It is also possible to obtain the remainder of the division of the value and the total number of physical pages after generation.

Furthermore, as a fourth method, at the time of system startup, "the value held by the hardware timer" and "the value obtained by adding all the erase counts (erase counts) of each block held by the flash memory control software" are multiplied. However, after generating a sufficiently large value with respect to the total number of pages of the flash memory being used, it is also possible to obtain the remainder of the division of the value and the total number of physical pages.
As a fifth method, at the time of system startup, "the value held by the hardware timer" and "the value held by the hardware real-time clock" are multiplied, and the total number of pages in the flash memory being used is calculated. Alternatively, after generating a sufficiently large value, the remainder of the division between the value and the total number of physical pages can be obtained.

  Subsequent to step S12, the physical address from which data is to be read and the offset value are obtained in the first block (S13). For example, if the value obtained in step S12 is “n + 2”, the first block is the second physical block. In the second physical block, a physical address “2 (= (n + 2) −n)” and an offset value “2” are obtained. Therefore, data is read from the physical address “2” in the second physical block, which is the first block to be read (S14).

  Subsequent to step S14, data reliability improvement processing (processing shown in detail in FIG. 4) such as error check processing is performed using the read data (S15). As described above, since the address of the error check page is obtained based on the time information acquired from the HW timer 22 every time the deterioration inspection system is activated, data is read from a random page every time the system is activated. It is possible to inspect and understand the deterioration state of the entire block. When step S15 ends, “i” is incremented to “i + 1” (S16).

  Following step S16, it is detected whether the total number of blocks has been exceeded with reference to the incremented “i” (S17), and if not, data is read from the physical address of the offset value in the i block (S18), Return to step S15. On the other hand, if it is detected in step S17 that it has been exceeded, the offset value is incremented by 1 (S19), and the process returns to step S14 to continue the processing.

  In FIG. 2, since the calculation result of step S12 is “n + 2”, the address “2” of the second block is set as the first read, and then the address “2” of the third block is set as the second read. To do. Here, when “i” is incremented to “i + 1”, “i” exceeds the total number of blocks, so the address “3” obtained by incrementing the offset value by 1 is used. For this reason, the address “3” of the first block, which is the first block, is the third read. Next to the first block, the process returns to the second block which is the first read block, and the address “3” is read for the fourth time, and then the address “3” of the third block is read for the fifth time. Here, when “i” is incremented to “i + 1”, since “i” exceeds the total number of blocks, the address “4” obtained by incrementing the offset value by 1 is used. According to the voice, the address “4” of the first block, which is the first block, is read out for the sixth time. Similarly, when data is read out one page at a time for each block, the process continues to move to the next.

  Thus, each time the deterioration inspection system is activated, the address of the error check page of the block to be read sequentially is obtained as an offset based on the time information acquired from the HW timer 22, so that the error check process for the same block is always performed. Without continuing, it becomes easy to read data from random pages for all blocks, and it is possible to inspect and grasp the deterioration state of the entire flash memory.

  Next, data reliability improvement processing (step S15) such as error check processing will be described in detail with reference to the flowchart shown in FIG. In step S15, the error bit number ES as a result of the error check is obtained (S21). Next, the first threshold value A set in advance is compared with the number of error bits ES (S22), and A <ES is detected (S23). If “YES” is determined in this step S23, all the pages of the block to which the page subjected to the error check belongs are refreshed (S24), and the process proceeds to a step S16. As a result, when the number of error bits ES per page is large and it can be predicted that an error has occurred in other pages in the same block, the entire data can be refreshed without error check to improve data reliability. it can.

  On the other hand, when branching to NO in step S23, the preset second threshold B (B <A) is compared with the number of error bits ES (S25), and B <ES is detected (S26). . If YES in step S26, the driver 12 obtains the number of erases and the number of reads related to the block to which the page belongs, which is stored in advance in the volatile memory 23, for example, and this number is shown on the graph of FIG. When plotted, it is detected whether it is above the deterioration threshold line TH shown in FIG. 5 (S27), and it is determined whether it is above (S28). As a result, if the process branches to YES, all the pages in the block are subjected to error checking (S29). The processing of each page in step S29 can perform steps S21 to S29 in the flowchart of FIG.

  This ensures that all pages in the block are not refreshed enough to be refreshed, but when it can be assumed that all pages need to be inspected, all pages are checked for errors to ensure data reliability. Improvements can be made. In step S27, on the deterioration threshold line TH shown in FIG. 5, a read count R corresponding to the erase count for the block to which the page belongs is obtained, and the read count for the block to which the page belongs is calculated as the read count. You may make it detect whether it is larger than the frequency | count R. FIG.

DESCRIPTION OF SYMBOLS 1 Software 2 Hardware 11 Data access software 12 Driver 21 Processor 22 HW timer 23 Volatile memory 24 Flash memory controller 25 Flash memory 31 Determination means 32 Processing means 33 Counting means

The flash memory deterioration inspection apparatus according to the present invention is a time counting unit capable of extracting time information, and a determination of determining a page on which error checking is performed by arithmetic processing based on the time information extracted from the time measuring unit. Means, a processing unit for designating a page determined by the determination unit, reading stored data and performing an error check, and a counting unit for obtaining the number of error bits at the time of the error check by the processing unit. ,
The processing means compares the number of error bits counted by the counting means with the first threshold value, and if the number of error bits is larger than the first threshold value, refreshes the block related to the error check, and further performs processing When the number of error bits is equal to or less than the first threshold, the means compares the number of error bits counted by the counting means with a second threshold smaller than the first threshold, and the number of error bits is less than the second threshold. If larger, the error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check .

The flash memory deterioration inspection method according to the present invention refers to a time measuring means capable of extracting time information, a time extracting step for extracting time information from the time measuring means, and an error check by arithmetic processing based on the obtained time information. A determination step for determining a page to be performed, a processing step for designating a page determined by the determination step, reading stored data and performing an error check, and an error bit number at the time of the error check by the processing step And a processing step compares the number of error bits counted in the counting step with a first threshold value, and if the number of error bits is greater than the first threshold value, the block relating to the error check In addition, in the processing step, the number of error bits is the first. When the number of error bits is less than the threshold, the number of error bits counted in the counting step is compared with a second threshold value that is smaller than the first threshold value. An error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block .

A program for inspecting deterioration of a flash memory according to the present invention is a device that designates a page for a flash memory in which data is stored, reads the stored data, performs an error check, and inspects the deterioration of the flash memory. The computer used is determined by referring to a clock from which time information can be extracted, time information is extracted, and a page determined by the determining means is determined based on the obtained time information, and a page on which error checking is performed by arithmetic processing. And processing means for reading out stored data and performing error checking, and functioning as counting means for obtaining the number of error bits at the time of error checking by the processing means . The processing means counts the error counted by the counting means. Compare the number of bits with the first threshold, and the number of error bits is less than the first threshold If the error bit number is less than the first threshold value, the processing unit refreshes the block related to the error check. Further, when the error bit number is equal to or less than the first threshold value, the processing unit counts the error bit number counted by the counting unit and the first threshold value. When a smaller second threshold value is compared and the number of error bits is larger than the second threshold value, all pages in the block related to the error check are based on the erase count and read count in the block related to the error check. It is characterized by performing an error check on .

Claims (12)

A time measuring means capable of extracting time information;
Time determining means for taking out time information from the time measuring means, and determining means for determining a page for performing error check by arithmetic processing based on the obtained time information;
A processing unit for designating a page determined by the determination unit, reading stored data, and performing an error check;
A deterioration inspection apparatus for a flash memory, comprising:   The processing means performs an error check by reading out stored data by sequentially specifying pages of the offset value in each block using the page determined by the determining means as an offset value. Flash memory deterioration inspection device. A counting means for obtaining the number of error bits at the time of an error check by the processing means,
The processing means compares the number of error bits counted by the counting means with a first threshold value, and when the number of error bits is larger than the first threshold value, refreshes the block related to the error check. The deterioration inspection device for a flash memory according to claim 1 or 2.   When the number of error bits is equal to or less than the first threshold value, the processing means compares the error bit number counted by the counting means with a second threshold value smaller than the first threshold value, and the error bit number is the second threshold value. 4. The flash memory according to claim 3, wherein if it is larger, an error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check. Deterioration inspection equipment. A time extraction step of referring to a time measuring means capable of extracting time information and extracting time information from the time measuring means;
A determination step for determining a page on which error checking is performed by arithmetic processing based on the obtained time information;
A processing step of designating a page determined by the determination step, reading stored data and performing an error check;
A method for inspecting deterioration of a flash memory, comprising:   6. The processing step performs read error checking by sequentially specifying the pages of the offset value in each block and reading the stored data using the page determined in the determination step as an offset value. Flash memory deterioration inspection method. It has a counting step for obtaining the number of error bits at the time of error checking by the processing step,
In the processing step, the number of error bits counted in the counting step is compared with a first threshold value, and if the number of error bits is larger than the first threshold value, the block related to the error check is refreshed. A method for inspecting deterioration of a flash memory according to claim 5 or 6.   In the processing step, when the number of error bits is equal to or less than the first threshold value, the number of error bits counted by the counting means is compared with a second threshold value smaller than the first threshold value, and the error bit number is the second threshold value. 8. The flash memory according to claim 7, wherein if it is larger, an error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check. Degradation inspection method. A computer used in an apparatus for inspecting deterioration of a flash memory by designating a page for the flash memory in which data is stored, reading the stored data, performing an error check,
A determination unit that extracts time information with reference to a clock from which time information can be extracted, and determines a page on which error checking is performed by arithmetic processing based on the obtained time information;
A processing unit for designating a page determined by the determining unit and reading stored data and performing an error check;
A program for inspecting deterioration of a flash memory characterized by functioning as   10. The processing unit performs an error check by reading out stored data by sequentially specifying pages of the offset value in each block using the page determined by the determining unit as an offset value. Flash memory deterioration inspection program. The computer is further caused to function as a counting means for obtaining the number of error bits at the time of error checking by the processing means,
The processing means compares the number of error bits counted by the counting means with a first threshold value, and when the number of error bits is larger than the first threshold value, refreshes the block related to the error check. The program for inspecting deterioration of a flash memory according to claim 9 or 10.   When the number of error bits is equal to or less than the first threshold value, the processing means compares the error bit number counted by the counting means with a second threshold value smaller than the first threshold value, and the error bit number is the second threshold value. 12. The flash memory according to claim 11, wherein if it is larger, an error check is performed on all pages in the block related to the error check based on the erase count and the read count in the block related to the error check. Degradation inspection program.

Images