JP2004071061A - Method and device for inspecting storage disk device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide stable quality by predicting the number of read errors of a sync byte (SB) caused by very small defects even without testing based on a data block length for each user regarding a method for inspecting a storage disk device. <P>SOLUTION: A writing/reading section 11 executes full-surface writing/reading based on ECC correction inhibition. A defective byte position detection section 12 obtains the position information of a read error occurrence position as the byte-from index (BFI) information of a defective byte, and stores the information in a defective byte position information storage section 13. A synch byte position calculation section 14 calculates the position information in the case of formatting based on a data block length expected to be used by a user as BFI information. A position information comparison section 15 compares the BFI information of a defective byte position with the BFI information of the position of each SB to obtain a coincident number. A nondefective/defective product determination section 16 determines a nondefective/defective product based on the coincident number. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for inspecting a magnetic disk device or the like, and particularly to a method for reading a sync byte (hereinafter referred to as SB) due to a defect without performing a test with a data block length corresponding to each user in a test process. The present invention relates to an inspection method and an inspection apparatus for a disk-type storage device, which predict the number of occurrences of a disk-type storage device and enable stable quality assurance.
[0002]
[Prior art]
In a shipping inspection of a magnetic disk, an inspection by writing and reading to and from a medium is performed. At that time, the test is performed under an environment that is several times more severe than the environment usually used by the user under temperature and voltage conditions that cause errors to occur intentionally.
[0003]
Further, in the test, in order to detect minute defects on the medium in advance, it is necessary to reduce not only the external environment but also the ECC (Error Correction Code) power added to the write data, that is, reduce the correction capability. Has been done. As a result, a minute defect that is corrected by the ECC is found, and if the number of defects is equal to or more than a certain reference, it is determined to be defective, so that a device of poor quality is not shipped.
[0004]
ECC power options at the time of testing include ECC correction prohibition (On the Fly correction prohibition) and ECC1 (1 burst correction) to ECC5 (5 burst correction). Many errors occur due to noise and the like, and the condition is far from the user environment (ECC4 or ECC5). Therefore, the defect rate of the current magnetic disk device is taken into consideration, and the ECC correction capability at the time of the test is minimum. However, ECC1 is often used. Therefore, in reading the entire surface of the medium (Read) during the test, extremely minute defects are ignored. The reason for ignoring these minute defects is that they have no effect on performance or life.
[0005]
The capacity of the magnetic disk device is represented by “(data block length) × (the number of data blocks of the entire device)”, but the data block length is usually 512 bytes at the user end, which is an industry standard value. However, as a function of the hard disk controller (HDC) mounted on the magnetic disk device, the data block length is allowed to be variable, and depending on the user, the data block length is not 512 bytes, which is the standard value, It may be used in the format of 512 ± α bytes.
[0006]
On the other hand, at the time of the shipping test, the data is tested in the format of 512 bytes, so if the format is used by the user at 512 ± α bytes (for example, the format is 520 bytes), Is tested with a data block length different from the data block length used in.
[0007]
[Problems to be solved by the invention]
If the magnetic disk device is used by the user with a data block length different from that at the time of the test, a very small defect ignored because the ECC1 is used as the minimum value of the ECC correction capability at the time of the test is synced. There is a possibility of hitting the byte (SB), and if the extremely minute defect hits the SB, an SB read error that did not occur during the test may occur.
[0008]
The SB is added to the data to synchronize the sectors, and is treated as uncorrected by the ECC, and thus has a redundancy. However, if a very small defect that has been neglected during the test just overlaps the SB, the SB read error occurrence rate may increase, and a read error that can be rescued by the user (an error that can be rescued by re-reading) frequently occurs. There was something. At this time, there is a low possibility that a serious read error such that data cannot be read will occur, but a rotation wait may occur due to an SB read error, causing a decrease in performance, and a complaint may be issued from the user.
[0009]
In order to eliminate the SB read error when the data block length used at the user is different from that at the time of the test, it is conceivable to perform the test with all the data block lengths that may be used by each user. The method is too costly and time consuming.
[0010]
SUMMARY OF THE INVENTION The present invention solves the above-mentioned problem and provides an inspection method of a disk storage device that can predict the number of SB read errors caused by defects without performing a test with a data block length corresponding to each user. The aim is to provide a method.
[0011]
[Means for Solving the Problems]
Hereinafter, a magnetic disk device will be described as an example of a disk type storage device. However, the present invention is not limited to a magnetic disk device, and is applicable to a test of a writable and readable disk type external storage device similar to a magnetic disk device. can do.
[0012]
In order to solve the above-mentioned problem, the present invention performs a test for extracting the position of a defective byte only in one data block length format, and performs calculations without actually formatting other data block length formats. The position of the SB is calculated, and the calculated position of the SB is compared with the position of the extracted defective byte to determine whether they match.
[0013]
Specifically, at the time of a shipping test with a data block length of 512 bytes, writing / reading of the entire surface is performed with ECC correction prohibited, and the position where a read error has occurred during data reading is obtained. This position is an error position including a minute defect that cannot be corrected at all, and is position information expressed as a distance (number of bytes) from a position (index) representing the head of one round of the track. The distance from this index is called a byte from index (BFI). At this time, all the BFI information at the error position is stored and stored in a memory or a medium (SA area).
[0014]
With the above operations, it is possible to determine the position on the medium where the minute defect exists by referring to the memory or the SA. However, the minute defect information detected by prohibiting the ECC correction at this time is not used at this time to judge a good / defective product of the apparatus.
[0015]
Next, the SB positions (SBn) of all the sectors on the track when the data is formatted with the data block length (512 ± α bytes) expected to be used by the user are calculated as BFI, and the position SBn and the previous work are obtained. The BFI information of the detected minute defect position is compared to check whether the position on the medium matches. As a result of the above comparison, the number of SB positions coincident with the minute defect positions is obtained. Therefore, it is possible to predict the number of SB read errors caused by minute defects when the data block length is (512 ± α bytes). It becomes possible.
[0016]
Therefore, if this method is used, the test is performed with only the standard data block length (512 bytes) in the test process, and even if the test is not performed with the block length tailored to each user, an SB read error due to a minute defect can occur. The quantity can be predicted, and if the predicted quantity is equal to or larger than the shipment reference value, it is determined that the product is defective, so that stable quality can be provided without shipping a device that may cause performance degradation at the user site. be able to.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, description will be made with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a system for realizing the present invention. In FIG. 1, a magnetic disk inspection device 10 includes a writing / reading unit 11, a defective byte position detecting unit 12, a defective byte position information storing unit 13, a sync byte position calculating unit 14, a position information comparing unit 15, a good / bad determining unit. 16 and is connected to a magnetic disk device 20 to be inspected.
[0018]
In the magnetic disk inspection device 10, the writing / reading unit 11 writes data to the entire surface of the magnetic disk device 20 having a standard data block length of 512 bytes, and then reads data from the entire surface. The ECC power at this time is ECC correction prohibited.
[0019]
The defective byte position detection unit 12 acquires a position where a read error has occurred during reading by the writing / reading unit 11 as BFI information of the defective byte position. At the same time, information such as a cylinder number, a head number, and the number of defective bytes is obtained. Here, the BFI information of the defective byte position is information of an error position including a minute defect that cannot be corrected at all, and is represented as a distance (the number of bytes) from a position (index) representing the head of one round of the track. Information (byte from index: BFI).
[0020]
The defective byte position information storage unit 13 stores the BFI information, the cylinder number, the head number, and the number of defective bytes of all the defective byte positions acquired by the defective byte position detection unit 12. The defective byte position information storage unit 13 is, for example, a memory or a system area (hereinafter, referred to as SA) on a medium. The position on the medium where the defect exists can be determined by referring to the defective byte position information storage unit 13. However, the defect information detected by prohibiting the ECC correction at this time is not used at this time to determine whether the magnetic disk device 20 is good or defective.
[0021]
The sync byte position calculation unit 14 determines the SB positions (hereinafter, referred to as SBn positions) of all the sectors on the track when the data is formatted with a data block length (512 ± α bytes in the present embodiment) which is assumed to be used by the user. Is calculated by each calculation, and this is used as the BFI information of the SBn position. The calculation of the SBn position can be performed by using an MCU (Micro Controller Unit) or the like.
[0022]
The position information comparing unit 15 compares and compares the BFI information of the defective byte position stored in the defective byte position information storage unit 13 with the BFI information of the SBn position calculated by the sync byte position calculating unit 14, and stores the information on the medium. Check if the positions match. As a result of the comparison, the number of SBn positions coincident with the defective byte position is obtained, so that it is possible to predict the number of SB read errors caused by minute defects when the data block length is 512 ± α bytes.
[0023]
Therefore, if this method is used, the test by actually writing and reading data in the test process is performed only with the standard data block length (512 bytes), and the test is not performed with the block length tailored to each user. Also, it is possible to predict the number of SB read errors that occur due to minute defects. The good / defective determining unit 16 determines that the predicted number is NG (defective) when the predicted number is equal to or larger than the shipping reference value, thereby preventing the user from shipping a magnetic disk device that may be degraded in performance. Stable quality can be provided.
[0024]
Hereinafter, the flow of the procedure in the present embodiment will be described. First, data is written to the entire surface of the medium with a data block length of 512 bytes, and subsequently, data is read from the entire surface with ECC correction prohibited. When a read error occurs, the write data and the read data are compared (compared) for each byte, and BFI information of a defective byte position, which is position information from the index of the error data (defective byte), is determined.
[0025]
FIG. 2 is a diagram showing an example of a procedure for determining BFI information of a defective byte position according to the present invention. First, as shown in FIG. 2A, a distance (the number of bytes) from an index (Index) to a servo (Servo) frame having error data (defective bytes) is calculated. Next, as shown in FIG. 2B, the distance (the number of bytes) to the head of a sector (Sector) having error data (defective bytes) in the corresponding servo frame in FIG. 2A is calculated. Subsequently, as shown in FIG. 2C, the distance (the number of bytes) from the head of the sector to the error data (defective byte) in the corresponding sector in FIG. 2B is obtained. For example, when there is error data (defective byte) in the data portion of FIG. 2C, the result is as shown in FIG.
[0026]
Therefore, the BFI information of the defective byte position is
BFI information of defective byte position = distance from index to corresponding servo frame (number of bytes)
+ Distance from the corresponding servo frame to the start of the corresponding sector (number of bytes)
+ Distance from the start of the sector to the defective byte (number of bytes)
It becomes.
[0027]
The BFI information obtained by the procedure shown in FIG. 2 is all recorded in a memory or an SA area on a medium. At this time, the cylinder number, the head number, the number of defective bytes, and the like are also recorded at the same time.
[0028]
Next, the SBn positions of all the sectors on the track when the medium is formatted with the data block length (data block length of 512 ± α bytes) expected to be used by the user are calculated. Here, the position information from the SBn position index is defined as the SBn position BFI information.
[0029]
FIG. 3 is a diagram showing an example of a procedure for calculating the BFI information at the SBn position in the present invention. First, the distance (the number of bytes) from the index to the servo frame having SBn is calculated (see FIG. 2A). Next, the distance (the number of bytes) from the servo frame to the head of a certain SBn is calculated (see FIG. 2B). Subsequently, as shown in FIG. 3, the distance (number of bytes) from the head of the sector to SBn is calculated in the corresponding sector.
[0030]
Therefore, the BFI information at the SBn position is
BFI information at SBn position = distance from index to corresponding servo frame (number of bytes)
+ Distance from the corresponding servo frame to the start of the corresponding sector (number of bytes)
+ Distance from the top of the sector to SBn (number of bytes)
It becomes.
[0031]
Next, the BFI information at the defective byte position and the BFI information at the SBn position are compared, and the number of both that matches is counted. At this time, it is assumed that SBn has a length of several bytes as the number of data, and in order to take into account rotation fluctuation, the defective byte position is expanded by several bytes before and after and compared.
[0032]
FIG. 4 is a diagram showing an example of comparison between BFI information at a defective byte position and BFI information at an SBn position in the present invention. FIG. 4A shows a case where the data block length is 512 bytes, and FIG. 4B shows a case where the data block length is 512 ± α bytes. The number of coincidences between the defective byte position and the SBn position is counted.
[0033]
Finally, the number of matches per head and the number of matches per unit are determined from the count of matches (the number of matches), the number of SB read errors per head when the data block length is formatted to 512 ± α bytes, The number of SB read errors per unit is predicted.
[0034]
From this result, it is possible to determine whether or not the magnetic disk device is likely to cause a performance drop at the user's site.
[0035]
FIG. 5 is a flowchart illustrating an example of a test process of the magnetic disk drive according to the present invention. In the shipping test, first, the data block length of the magnetic disk device to be inspected is set to 512 bytes, and data writing and reading are performed on the entire surface by the ECC 1 (step S10). If there is a read error, the position of the defective byte is obtained by comparing the written data with the read data and counted as a defect (defect) (step S11). It is determined whether or not the counted number of defects is equal to or greater than a reference value (step S12). If the number is equal to or greater than the reference value, the magnetic disk device to be inspected is determined as a defective device.
[0036]
If the value is not equal to or greater than the reference value in the process of step S12, data writing and reading are performed over the entire surface without ECC correction while the data block length of the magnetic disk device to be inspected remains at 512 bytes (step S13). If there is a read error, the defective byte position is determined by comparing the written data with the read data, and the BFI information of the defective byte position is obtained (step S14). The acquired BFI information of the defective byte position is recorded in a memory, a SA on a medium, and the like together with information such as a cylinder number, a head number, and the number of defective bytes (step S15).
[0037]
The position of the SB (referred to as SBn) when formatted with the data block length (512 ± α bytes) assumed to be used by the user is calculated by one track calculation, and the position information from those indexes is used as the SBn position information. This is set as BFI information (step S16). The BFI information at the SBn position of each sector is compared with the BFI information at the defective byte position to determine whether they match (step S17). If they match, the number of matches is counted (step S18), and if they do not match, the number of matches is not counted.
[0038]
It is determined whether the comparison has been completed for all cylinders and all heads (step S19). If the comparison has not been completed, the track to be compared is changed (step S20), and the process returns to step S16.
[0039]
If the comparison has been completed for all cylinders and all heads in the determination in step S19, the number of matches per head and the number of matches per device are determined (step S21). It is determined whether the number of matches per head exceeds a predetermined standard value (step S22). If it exceeds, the magnetic disk device to be inspected is determined as a defective device. Similarly, it is determined whether the number of matches for the entire apparatus exceeds a predetermined standard value (step S23). If the number exceeds, the magnetic disk device to be inspected is determined as a defective device. If both the number of matches per head and the number of matches per device do not exceed the predetermined standard values, the magnetic disk device to be inspected is regarded as a non-defective device.
[0040]
The features of the present embodiment are listed below.
[0041]
(Supplementary Note 1) In a disk-type storage device inspection method for checking whether a disk-type storage device is good or defective,
A process of writing data to the entire surface of the disk-type storage device with a data block length of X bytes, and reading the entire data without ECC correction;
Calculating the position where the reading error has occurred as defective byte position information represented by the number of bytes from the index which is the head of one round of the track, and storing the calculated defective byte position information;
The sync byte position information expected when the disk-type storage device is formatted with a data block length Y bytes different from the data block length X bytes by the number of bytes from the index which is the head of one round of the track. A process of calculating as position information by calculation,
Comparing the defective byte position information with the sync byte position information to see if they match.
[0042]
(Supplementary note 2) In the inspection method for a disk-type storage device according to supplementary note 1,
The number of coincidences between the defective byte position information and the sync byte position information is calculated for each head or for the entire apparatus, and an inspection target is determined based on whether a predetermined threshold per head or a predetermined threshold for the entire apparatus is exceeded. A method of inspecting a disk-type storage device according to claim 1, wherein the determination is made as to whether the disk-type storage device is good or defective.
[0043]
(Supplementary Note 3) In a disk-type storage device inspection apparatus for checking whether a disk-type storage device is a good product or a defective product,
Writing / reading means for writing data to the entire surface of the disk-type storage device with a certain data block length X bytes and reading the entire data without ECC correction;
Defective byte position detecting means for calculating the position where the reading error has occurred as defective byte position information represented by the number of bytes from the index which is the head of one round of the track;
Defective byte position information storage means for storing the defective byte position information;
Sync byte, which is the position information of the expected sync byte when the disk-type storage device is formatted with the data block length Y byte different from the data block length X byte, expressed by the number of bytes from the index which is the head of one round of the track. Sync byte position calculating means for calculating by calculation as position information;
An inspection apparatus for a disk-type storage device, comprising: position information comparing means for comparing whether the defective byte position information and the sync byte position information match.
[0044]
(Supplementary note 4) In the inspection device for a disk-type storage device according to supplementary note 3,
The number of coincidences between the defective byte position information detected by the position information comparing means and the sync byte position information is calculated for each head or for the entire apparatus, and a predetermined threshold per head or a predetermined threshold for the entire apparatus is calculated. An inspection apparatus for a disk-type storage device, comprising: a good / defective determining unit for determining whether the disk-type storage device to be inspected is a non-defective product or a defective product based on whether or not the number exceeds a predetermined value.
[0045]
(Supplementary Note 5) A disk-type storage device inspection program for inspecting whether a disk-type storage device is good or defective.
A process of writing data to the entire surface of a disk storage device with a data block length of X bytes and reading the entire data with ECC correction prohibited;
Calculating the position where the reading error has occurred as defective byte position information represented by the number of bytes from the index which is the head of one round of the track, and storing the calculated defective byte position information;
Sync byte, which is the position information of the expected sync byte when the disk-type storage device is formatted with the data block length Y byte different from the data block length X byte, expressed by the number of bytes from the index which is the head of one round of the track. Processing to calculate by calculation as position information;
Comparing the defective byte position information with the sync byte position information to see if they match.
An inspection program for a disk storage device to be executed by a computer.
[0046]
【The invention's effect】
As described above, by using the present invention, in the shipping test, the test is performed only with the standard data block length (512 bytes), and even if the test is not performed with the data block length according to the user, the SB read error due to the defect can be performed. It is possible to provide a stable quality disk-type storage device without shipping a disk-type storage device that may cause performance degradation to the user.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration example of a system that implements the present invention.
FIG. 2 is a diagram showing an example of a procedure for determining BFI information of a defective byte position according to the present invention.
FIG. 3 is a diagram showing an example of a procedure for calculating BFI information at an SBn position according to the present invention.
FIG. 4 is a diagram illustrating an example of comparison between BFI information at a defective byte position and BFI information at an SBn position according to the present invention.
FIG. 5 is a flowchart illustrating an example of a test process of the magnetic disk drive according to the present invention.
[Explanation of symbols]
Reference Signs List 10 magnetic disk inspection device 11 writing / reading unit 12 defective byte position detecting unit 13 defective byte position information storage unit 14 sync byte position calculating unit 15 position information comparing unit 16 good / bad determination unit 20 magnetic disk device

Claims (3)

In a disk-type storage device inspection method for checking whether a disk-type storage device is good or defective,
A process of writing data to the entire surface of the disk-type storage device with a data block length of X bytes, and reading the entire data without ECC correction;
Calculating the position where the reading error has occurred as defective byte position information represented by the number of bytes from the index which is the head of one round of the track, and storing the calculated defective byte position information;
The sync byte position information expected when the disk-type storage device is formatted with a data block length Y bytes different from the data block length X bytes by the number of bytes from the index which is the head of one round of the track. A process of calculating as position information by calculation,
Comparing the defective byte position information with the sync byte position information to see if they match. The inspection method for a disk-type storage device according to claim 1,
The number of coincidences between the defective byte position information and the sync byte position information is calculated for each head or for the entire apparatus, and an inspection target is determined based on whether a predetermined threshold per head or a predetermined threshold for the entire apparatus is exceeded. A method of inspecting a disk-type storage device according to claim 1, wherein the determination is made as to whether the disk-type storage device is good or defective. In a disk-type storage device inspection device for checking whether a disk-type storage device is good or defective,
Writing / reading means for writing data to the entire surface of the disk-type storage device with a certain data block length X bytes and reading the entire data without ECC correction;
Defective byte position detecting means for calculating the position where the reading error has occurred as defective byte position information represented by the number of bytes from the index which is the head of one round of the track;
Defective byte position information storage means for storing the defective byte position information;
The sync byte position information expected when the disk-type storage device is formatted with a data block length Y bytes different from the data block length X bytes by the number of bytes from the index which is the head of one round of the track. Sync byte position calculating means for calculating by calculation as position information;
An inspection apparatus for a disk-type storage device, comprising: position information comparing means for comparing whether the defective byte position information and the sync byte position information match.

Images