JPH04353673A - Data read-write method of disk storage device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention relates to a method for reading and writing data in an optical disk type or magnetic disk type disk storage device, and more precisely, to a disk storage device in which the inside of the disk is divided into sectors each having a repeated synchronization code at the beginning. Concerning how to read and write data.

0002

2. Description of the Related Art In a disk storage device such as the above-mentioned optical disk, a large number of tracks are set on the disk, and each track is divided into a plurality of sectors, which are data recording units. However, when reading and writing data to each sector, it is necessary to precisely synchronize the operation of the related circuitry to the position where the code representing the data is written, so a synchronization code is placed at the beginning of each sector in advance. It is customary to write the data code and read it to synchronize the operation of related circuits with the position of the data code. Hereinafter, an example of code allocation within this sector will be conceptually explained with reference to FIG. 2.

FIG. 2 shows a track T set on the disk surface.
FIG. 2 is an enlarged developed view of one sector S in FIG. As shown in the figure, a plurality of synchronization codes SC are written in succession at the beginning of each sector S, then an area for writing data codes DC is set, and finally a gap G is set to provide some margin. is placed. The synchronization code SC is usually a very simple code, for example, 001, and this is repeatedly written several tens to a hundred times or more. The data code DC area includes an area for the track and sector verification data ID and the data body DT, and in front of these are areas for the verification mark AM and data mark DM, respectively, and finally an error correction code EC. There will be an area for Data body D
Of course, the largest area, for example 1024 bytes, is allocated to the area for T.

Next, an example of a conventional method for reading and writing data using the synchronization code SC placed at the beginning of the sector S will be described with reference to FIG. The figure shows the configuration of the main parts of a disk storage device, and the disk 1 is assumed to be, for example, a phase change type optical disk.

The disk 1 is driven by a spindle motor 2 at a constant speed specified by a clock pulse CP received by a drive circuit 3. The data read/write head 4 is positionally operated by a mechanism not shown, and irradiates a spot of laser light L onto a predetermined track within the surface of the disk 1. A light emitting drive circuit 5 attached to this head 4 has a read/write command R.
A light emitting element such as a laser diode in the head 4 emits light with a light intensity corresponding to W and a write signal Sw, and a read signal amplifier 6 detects light from a photodiode or the like in the head 4 that receives reflected laser light from the disk 1. The detection output of the element is amplified and output as an analog read signal Sr.

[0006] A code demodulation circuit 7 receiving the read signal Sr binarizes the analog waveform and converts it into a digital code signal Sc. Further, the data separation circuit 10 receives this code signal Sc and causes the attached PLL circuit 11 to establish a synchronization pulse SP synchronized with the read portion of the synchronization code SC included in the code signal Sc, and based on this, the data separation circuit Separate the reading part of the code DC and use it as the data code signal Sd as the synchronization pulse SP
Output with.

In this example, the read/write means 20 consists of an encoder/decoder circuit 21 and a data conversion means 22, the former decoding the data code signal Sd into a coded signal of a predetermined method in synchronization with the synchronization pulse SP, and the latter decoding the data code signal Sd into a coded signal of a predetermined method. give to The data conversion means 22 is a small processor, which converts this serial coded signal into parallel digital data and stores it in the RAM 22a, or vice versa.
The data in 22a is converted into a serial coded signal, encoded into a write signal Sw in the encoder/decoder circuit 21, and then provided to the light emitting drive circuit 5 described above.

A processor 30 is built into the disk storage device for data control and its internal control, and is connected to a host computer (not shown) via an internal bus 31, an interface circuit 32, and an external bus 33. Alternatively, a clock pulse CP is supplied from the attached pulse generation circuit 34 to the spindle motor drive circuit 3 and data separation circuit 10, and a read/write command RW is issued based on a command from the computer. The read/write means 20 operates in conjunction with the processor 30 via a communication bus 35, etc., and transfers write data from the internal bus 31 to the RAM 2 based on the command.
2a and outputs it as a write signal Sw, or receives the data code signal Sd and temporarily stores the read data represented by it in the RAM 22a and then puts it on the internal bus 31, and transfers the temporarily stored data to it. It is checked using the included error correction code EC, and if necessary, the contents are corrected and then the processor 30 is contacted.

In the disk storage device configured as shown in FIG. 3 described above, each sector S
As described above, the data separation circuit 10 generates the synchronization pulse SP based on the synchronization code SC at the beginning of the synchronization code SC, and the read/write means 20 synchronizes with this to read the data code signal Sd and accurately convert it into data. Furthermore, when data is written, it is common practice to perform a read verification that immediately reads and checks the data to ensure the accuracy of the write. Furthermore, even if reading fails once when reading data, it is customary to continue reading and trying up to three times, for example, to perform so-called re-reading.

0010

[Problems to be Solved by the Invention] Even in the conventional read/write method of a disk storage device, as mentioned above, reading verification is performed when writing data, and when reading data, the reading operation is synchronized with a synchronization code, and if necessary, Rereading improves the reliability of read/write operations and prevents read/write errors from occurring, but if the disk recording medium is originally defective or scratched during use, it may not be possible to completely eliminate errors. There are problems that cannot be prevented.

[0011] Such read/write errors actually occur during read verification after data writing or when data read during data reading is checked sector by sector using the error correction code EC as described above, but the cause of this error will be explained in detail. After tracking and analyzing, it was found that this problem is particularly likely to occur when there is a defect or scratch on the recording medium where the synchronization code SC is written. That is, the area of the data code DC in FIG. 2, especially the data body DT.
Even if there are some defects in the area, the error correction code EC
However, if there is a defect in the area of the synchronization code SC, correction using the error correction code EC will of course not work, so the timing of the synchronization pulse SP will be incorrect and the data read in synchronization with it will be lost. is disturbed to the extent that it cannot be corrected even by the error correction code EC.

[0012] Furthermore, what is troublesome about the occurrence of read/write errors is that errors may or may not occur depending on the usage environment such as temperature. That is, if there is a defect or scratch on the recording medium where the synchronization code SC is written, but the degree of the defect is not that large, for example, even if the analog waveform of the read signal Sr in FIG. This is because a correct code signal Sc may be reproduced during binarization by , and an error does not necessarily occur at a location where there is a defect in the medium. Therefore, if by chance no error is detected during read verification immediately after data is written and no write error is detected, an error will be detected during subsequent data reading and the data will not be read correctly no matter how many times you try to reread it. Situations may occur.

[0013] Depending on the disk storage device, the
For example, the data mark DM other than the data body DT in the data code DC may not be corrected by the error correction code EC, and if there is a defect in the medium in that area, an error will occur as well. , synchronization pulse SP
If the code is normal, it is often possible to read the code by slightly relaxing the criteria for determining the code pattern. Compared to this, there is a much greater possibility that a defect in the medium in the synchronization code SC area will cause an error.

[0014] The present invention solves such conventional problems, and
An object of the present invention is to provide a reading/writing method in which there is little risk of occurrence of reading/writing errors even if there is a defect in a recording medium in a synchronization code area placed at the head of each sector of a disk storage device.

[0015]

[Means for Solving the Problems] The above object is to provide a means for detecting that a predetermined number or more of synchronization codes are included in a code signal read out from the recorded contents of each sector and emitting a detection signal; In addition, in the first method of the present invention, when writing data, the masking state for the data code signal is canceled by the detection signal of the synchronization code after writing to each sector. On the other hand, the second method of the present invention synchronizes the read verification after data writing and the data reading by performing reading verification of the written data and reading the data under the condition that the mask state for the data code signal is released when reading the data. This is achieved by re-reading the data in sectors where data reading has failed, under the conditions that the masked state for the data code signal is released by the encoded code detection signal, and that the masked state for the data code signal is released from the sector where data reading has failed. Ru.

Note that it is advantageous for the synchronization code detection means in the above structure to be attached to or incorporated into a conventional data separation circuit, and for the mask means, a single logic gate such as an AND gate is used. You can configure it.

In both the first and second methods, it is preferable that the detection signal by the synchronization code detection means be generated on the condition that a predetermined number or more of synchronization codes are consecutively included in the code signal. If the read verification after writing data fails, it is preferable to write the data for that sector in another sector, and it is desirable to perform the read verification in this case without correction by an error correction code. Furthermore, in the second method, if data reading once fails, even if rereading succeeds, it is desirable to write the data in that sector to another sector.

[0018]

[Operation] As described in the configuration in the previous section, in the present invention, a synchronization code detection means is provided to detect that a predetermined number of synchronization codes, for example, eight or more, are included in the code signal read out from each sector. In addition to confirming that a substantial portion of the recording medium in the synchronization code area is healthy and the synchronization pulse has been normally established, a masking means is provided to mask the data code signal from which data following the synchronization code is read. After that, when data is written, the masked state for the data code signal is released only by a detection signal indicating that the synchronization code of that sector has been normally detected when reading the data of each sector after writing. By doing this under these conditions, we minimize the possibility of a hidden write error occurring, such as when a read verification is accidentally successful despite a defect or scratch on the recording medium in the synchronization code area and a read error occurs later. Exclude.

[0019] If the synchronization code of a certain sector is not detected normally, the data code signal of that sector will be masked, so reading verification will naturally fail, and even if you notify the host computer of this fact, it will not work. However, it is most rational to rewrite the data to another healthy sector.

When reading data, in the first method, the data in each sector is read under the condition that the mask state for the data code signal is released, so that if the medium in the synchronization code area is damaged after data is written, The second method also makes it easier to successfully read or reread data and reduces the probability of reading errors.The second method is performed under strict conditions in which the mask state for the data code signal is released by the detection signal of the synchronization code. Then, the probability of occurrence of a read error is reduced by re-reading the sector in which data reading has failed under relaxed conditions in which the masked state for the data code signal is released. In the second method, if the first reading fails, it is obvious that there is a scratch on the medium, so it is desirable to rewrite the data in another healthy sector for safety.

[0021]

[Embodiment] An embodiment of the present invention will be described below with reference to FIG. FIG. 1 is a block diagram of the main parts of a disk storage device suitable for implementing the data read/write method of the present invention, and parts corresponding to those in FIG. . It is assumed that the configuration of the sector S in the disk 1 of FIG. 1 is as shown in FIG. 2, and that the illustrated disk storage device is a phase change type optical disk storage device as in FIG. 3. Note that the sector S shown in FIG.
The configuration is common; in the case of an optical disc, the verification code ID is provided in a location other than the sector, so this and the verification mark AM are omitted, and the data code DC starts from the data mark DM. Furthermore, in the case of an optical disc, the synchronization code SC is usually called VFO and the data mark DM is called SYNC.

To implement the present invention, as shown in FIG. 1, synchronization code detection means 12 and masking means 40 are provided in addition to the conventional configuration of FIG. The synchronization code detection means 12 is preferably provided in association with the data separation circuit 10 as shown in the figure, and is provided with a code signal Sc read out from the recorded contents of the disk 1 to detect synchronization of a predetermined number or more. It is monitored whether or not the read portion of the code SC is included, and a detection signal DS is generated when, for example, eight or more synchronization codes SC are detected in succession.

In this embodiment, the masking means 40 is constituted by a single AND gate, one input of which receives the data code signal Sd from the data separation circuit 10, and the other input receives the above-mentioned detection signal DS through an OR gate 41. and provides the output to the encoder/decoder circuit 21 in the reading/writing means 20. The OR gate 41 receives a detection signal DS from the synchronization code detection means 12 at one input thereof.
and receives a mask release signal MS from the processor 30 at the other input. Since the mask release signal MS is always kept in a low state, the synchronization code detection means 1
2 detects eight or more synchronization codes SC and sends a detection signal DS.
Unless Sd is set to a high state, the AND gate for the masking means 40 is in a disabled state and serves to mask the data code signal Sd and reserve its transmission to the read/write means 20.

In the disk storage device of FIG. 1 configured as described above, the read/write operation is performed by the read/write command R from the processor 30 based on the command from the host computer.
Specified by the logical state of W. When data writing is commanded, the read/write means 20 uses the internal bus 3 as described above.
The write data taken into the RAM 22a of the data conversion means 22 from the encoder/decoder circuit 2
1 to the light emitting drive circuit 5 as a write signal Sw to cause the head 4 to write this data into a predetermined sector S in the disk 1. In this embodiment, the synchronization code SC
In order to accurately maintain the writing positional relationship between the data code DC and the data code DC, the entire sector S shown in FIG. 2 is written.

For this reason, as shown in FIG. 1, a read/write command RW and a clock pulse CP are given to the data separation circuit 10 from the processor 30 side, and when the read/write command RW instructs writing, a clock pulse is sent as a synchronization pulse SP. The pulse CP is directly outputted from the data separation circuit 10, and the read/write means 20 that receives it generates a write signal Sw in synchronization with this clock pulse CP.

Immediately after writing this data, the process moves to its read verification, and after setting the read/write command RW to a read designation state, the data just written is read from the same sector. At this time, a synchronization pulse SP based on the synchronization code SC is applied from the data separation circuit 10 to the read/write means 20, but in the present invention, reading verification is performed with the mask release signal MS in the same state as when writing. A predetermined number of synchronization codes SC from the synchronization code detection means 12
The data code signal Sd is masked by the masking means 40 unless the synchronization code detection signal DS indicating that the synchronization code detection signal DS is detected is generated.

When the synchronization code detection signal DS becomes high, the data code signal Sd is given to the read/write means 20, so the data conversion means 22 temporarily stores the data read in synchronization with the synchronization pulse SP in the RAM 22a and then reads the data. Then, the content of the data body DT is checked using the error correction code EC included in it, and the result is sent to the processor 3.
0 via the communication bus 35.

At this time, if there is a defect or scratch on the medium in the area for synchronization code SC in sector S, synchronization code detection signal D
S is not generated and the data code signal Sd is read/written by the read/write means 2.
Therefore, even if the synchronization pulse SP from the data separation circuit 10 is normal, the reading verification will fail in the method of the present invention. Furthermore, in order to make this reading verification more rigorous, when the synchronization code detection signal DS is generated and data is read, when the data conversion means 22 checks the contents, the error correction code EC is
It is desirable to use only the check function to prevent data contents from being corrected.

If the read verification is successful, the data writing is of course completed, but if it is unsuccessful, in this embodiment, the data is written to another healthy sector without notifying the computer of this fact. At this time, the processor 30 that has received the report of failure specifies another or a spare sector and performs replacement processing for this sector by repeating data writing until the reading and verification is successful.

[0030] The data reading operation is the first and second one of the present invention.
Different in ways. In the first method, in order to reduce the probability of occurrence of a reading error as much as possible, the processor 30 supplies a high mask release signal MS to the OR gate 41 and the masking means 4
Data reading is performed with the AND gate for 0 enabled and the mask for the data code signal Sd released. That is, even if, for example, the medium in the area for synchronization code SC is damaged after data is written and the synchronization code detection signal DS is not generated, the first method provides the data code signal Sd to the read/write means 20 regardless of this. Read the data.

In this data reading, read/write command R
Of course, the data is read into the data conversion means 22 with W set to read, the content of the data is checked using the error correction code EC, and errors are corrected if necessary. If this first reading fails, rereading is repeated, for example, three times with the mask removed, and only if all of these readings fail is a reading error, but unless the scratches on the medium are particularly large, such an abnormal situation will not occur. Does not occur.

In the second method of the present invention, similarly to the reading detection after writing, data reading is performed using the synchronization code detection signal DS for the data code signal Sd while keeping the mask release signal MS in the low state. This is carried out under the condition of canceling the masked state, and if this first data reading fails, the masking canceling signal MS is switched to high, and data rereading is repeated about three times, for example, under the condition of canceling the masking of the data code signal Sd. As you can see, in the second method, the probability of success in reading data the first time is slightly lower than in the first method, but since the probability of success in re-reading is the same, the overall probability of reading errors occurring is almost unchanged. .

However, in this second method, if there is a scratch on the medium that would impede reading of the synchronization code SC, it can be reliably detected during the first reading, so even if re-reading is successful, it is safe. Therefore, if the data is rewritten to another healthy sector, the probability of a read error occurring can be further reduced by excluding the damaged sector each time data is read. Note that even with the first method, it is possible to distinguish whether the data was successfully read the first time or when it was reread, so in the latter case the data may be rewritten to another sector, but if the first reading fails, the Since this often occurs due to causes other than scratches, it is more reasonable to use the second method for rewriting data to another sector.

The method of the present invention is not limited to the embodiments described above, but can be implemented in various embodiments. For example, although the synchronization code detection means 12 is provided in conjunction with the data separation circuit 10 in the embodiment, it is actually advantageous to incorporate it therein.
If the data separation circuit 10 itself has this function, it is only necessary to utilize this function. Since the masking means 40 and the attached OR gate 41 can be very simple logic gates, it is convenient to incorporate them into the data separation circuit 10. Furthermore, if the synchronization code detection signal DS is given to the processor 30, the operation of the OR gate 41 can be taken over by the software, and the mask release signal MS can be given directly to the masking means 40. Furthermore, the configuration of the disk storage device in FIG. 1 is merely an example, and in reality, various configurations may be adopted.

[0035]

As described above, the present invention includes means for detecting that a predetermined number or more of synchronization codes are included in a code signal read out from the recorded contents of each sector and issuing a detection signal to the disk storage device; and means for masking the data code signal from which the data has been read, and in the first method, when writing the data, reading and verifying the written data is performed under the condition that the masked state for the data code signal is canceled by the synchronization code detection signal after writing, When reading data, the data is read under the condition that the mask for the data code signal is released, and in the second method, the mask state for the data code signal is released by synchronizing the read verification after data writing and the data reading. The following effects can be obtained by re-reading data under the conditions that the data code signal is unmasked for sectors where data reading has failed.

(a) During reading verification after data writing, only if it is confirmed that the code signal from which the data has been read contains a predetermined number or more of synchronization codes, synchronization is achieved by canceling the mask for the data code signal. It almost certainly prevents the occurrence of hidden write errors, where a read verification is accidentally successful despite defects or scratches on the media in the encoded area, and a read error later occurs. The reliability of the data including the synchronization code to be written can be significantly improved compared to the past.

(b) When reading data, the first method reads the data under the condition that the data code signal is unmasked, and the second method reads the data under the condition that the data code signal is unmasked by the synchronization code detection signal. By first reading the data and re-reading it under the condition that the mask for the data code signal is released in case of failure, even if the medium is scratched after data has been written, the data can be read easily and the written data can be easily read. Combined with high reliability, the probability of reading errors occurring can be minimized.

(c) If the read verification after data writing or the first data reading in the second method fails, the data is rewritten to another healthy sector, thereby preventing defects or scratches on the medium each time data is read or written. By eliminating certain sectors, the data recording state of the disk storage device can be kept in the best state over a long period of time.

As described above, by applying the present invention to an optical disk type or magnetic disk type disk storage device, even if a defect or scratch occurs on the recording medium or occurs during use, the media area for the synchronization code can be saved. It is possible to prevent data from being written to a defective sector and to minimize the probability of read errors, thereby further improving the long-term operational reliability of the disk storage device.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram illustrating the configuration of main parts of a disk storage device to which a data read/write method according to the present invention is applied.

FIG. 2 is a partially enlarged exploded view of one sector on the disk surface showing an example of the structure of a sector in which data including a synchronization code handled by the method of the present invention is recorded.

FIG. 3 is a diagram illustrating a main part of a disk storage device using a conventional disk read/write method.

[Explanation of symbols]

1 disk 4 read/write head 7 read signal code demodulation circuit 10
Data separation circuit 12 Synchronization code detection means 20
Reading/writing means 30 Embedded processor of the disk storage device 40 Masking means or AND gate D
C Data code DS Synchronization code detection signal MS
Mask release signal RW Read/write command S Sector SC Synchronization code Sc Read data code signal Sd
Data code signal SP Synchronization pulse Sr Read signal Sw Write signal

Claims (5)

[Claims] 1. A method for reading and writing data in a disk storage device in which the disk is divided into sectors having a repeated synchronization code at the beginning, the code signal from which the recorded content of each sector is read includes a synchronization code. A means for detecting that a predetermined number or more is included and issuing a detection signal, and a means for masking the data code signal read out from the data in the sector following the synchronization code are provided. Reading and verifying write data is performed for each sector under the condition that the mask state for the data code signal is released by the detection signal,
A method for reading and writing data in a disk storage device, characterized in that when reading data, data in each sector is read under conditions in which a masked state for a data code signal is released. 2. A method for reading and writing data in a disk storage device in which a disk is divided into sectors having a repeated synchronization code at the beginning, wherein the code signal read out the recorded contents of each sector includes a synchronization code. A means for detecting that a predetermined number or more is included and issuing a detection signal, and a means for masking a data code signal that reads data in a sector following a synchronization code, are provided, and read verification after data reading and writing is provided. This is done under the condition that the masked state for the data code signal is canceled by the detection signal of the synchronization code, and for sectors where data reading has failed, data is reread under the condition that the masked state for the data code signal is canceled. A data read/write method for a disk storage device, characterized in that: 3. The method according to claim 1 or 2,
1. A method for reading and writing data in a disk storage device, characterized in that a detection signal is generated on the condition that a predetermined number or more of synchronization codes are consecutively included in a code signal. 4. The method according to claim 1 or 2,
A method for reading and writing data in a disk storage device, characterized in that data for a sector that fails read verification after writing data is written to another sector. 5. A method for reading and writing data in a disk storage device according to claim 2, wherein the data is written in another sector after re-reading the data.