JP2004220068A - Memory card and method for writing data in memory - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem with writing on a flash memory wherein writing on a flash memory where writing errors are likely to occur, such as a multivalued flash memory, often results in writing errors, while defective blocks are likely to be generated in large numbers. <P>SOLUTION: In the event of a writing error (step 4), writing is retried on another block. If a writing error occurs again, a writing process is ended because of the error (step 7), while data are read and an error check is performed (step 8). Only when there are uncorrectable errors, the block is registered as a defective block (step 9); when there are correctable errors, it is not registered as a defective block. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a memory card and a method of writing data to a memory, which are characterized by error processing when writing data to a rewritable nonvolatile memory.
[0002]
[Prior art]
In recent years, as storage devices for storing digital information, memory cards and devices equipped with rewritable nonvolatile memories have become widespread. In particular, since memory cards are small and have excellent earthquake resistance, they are expected to be increasingly used in portable devices in the future.
[0003]
A typical rewritable nonvolatile memory is a NAND flash memory (hereinafter referred to as a flash memory). Whether the flash memory is used or not used is managed in block units. For example, “1” is written as management data to an unused block, and when data is written, the management data is changed to “1”. Rewritten from "1" to "0".
[0004]
In general, flash memory rarely does not normally write. For this reason, the flash memory has a function of reading (reading) a status indicating whether the writing was normal or abnormal, and by reading this status after writing the data, the writing is normally performed. It is necessary to determine whether this has been done. When a write error occurs, the management data of the corresponding block is rewritten from "1" to "0" so that the block is not selected as a write block again.
[0005]
As described above, regarding the write error of the flash memory, the one described in Patent Document 1 is known.
[0006]
[Patent Document 1]
JP-A-2002-108720 (paragraphs 0003 to 0004)
[0007]
[Problems to be solved by the invention]
In the conventional flash memory, binary data of "0" or "1" is written in one cell, but in recent years, four values of "00", "01", "10" or "11", or A flash memory for writing the above multi-valued data has been developed, and some of the flash memories have been put into practical use. A multi-valued flash memory can be configured with a smaller area than a binary flash memory having the same capacity and can reduce the cost of the flash memory. It is considered something. However, a multi-valued flash memory has a higher probability of causing an error during writing than a conventional binary flash memory. Therefore, if writing is performed in the conventional manner, a writing error frequently occurs, and a large number of defective blocks may be generated.
[0008]
SUMMARY An advantage of some aspects of the invention is to provide a memory card and a method of writing data to a memory that can reduce the frequency of write errors and reduce the generation of defective blocks.
[0009]
[Means for Solving the Problems]
In order to solve this problem, a memory card according to the present invention is a memory card equipped with a rewritable nonvolatile memory, and includes a management table for managing a data write state of the nonvolatile memory, and a block unit for the nonvolatile memory. Means for writing data, determining means for determining whether the data writing has been performed normally, and rewriting the block on which the writing has been performed when the determining means determines that the data writing has been performed normally. Means for updating the management table so as to prohibit writing and terminating normally, and means for retrying data writing to a different block of the nonvolatile memory when the judging means judges that data writing was not executed normally. Determined that the data writing by retry was not executed normally by the determining means Error correction means that reads data from the block that was written at the time of retry and corrects the error. A means for updating the table and terminating the error without determining a bad block when the error is determined to be within the correction capability is provided.
[0010]
As a result, the frequency of occurrence of error termination can be reduced by performing the retry process, and an increase in the number of defective blocks can be prevented by setting a defective block only when an error cannot be corrected.
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
The invention according to claim 1 of the present invention is a memory card equipped with a rewritable nonvolatile memory, wherein the management table manages a data write state of the nonvolatile memory, and the nonvolatile memory is configured based on the management table. Means for writing data to the memory in block units, determining means for determining whether or not the data writing has been performed normally, and when the determining means determines that the data writing has been performed normally, Means for updating the management table so as to prohibit rewriting of the block in which the data has been written and terminating the data normally, and when the determination means determines that the data writing has not been performed normally, Means for retrying data writing to a different block, and data writing by said retry When it is determined by the determination unit that the execution was not performed normally, an error correction unit that reads data from the block that was written at the time of the retry and performs error correction, and an error that exceeds the correction capability by the error correction unit. A memory card having means for updating the management table with a block used for retry as a bad block only when it is determined, and terminating an error without determining a bad block when the error is within the correction capability.
[0012]
According to a second aspect of the present invention, there is provided a method for writing data to a rewritable nonvolatile memory, the method comprising: writing data to the nonvolatile memory based on a management table for managing a data write state of the nonvolatile memory; A step of writing data in block units, a determining step of determining whether or not the data writing has been normally performed; and a step of determining whether the writing of the data has been normally performed by the determining step. Updating the management table so as to prohibit rewriting of the performed block and terminating the block normally; and, when it is determined by the determination step that the data writing has not been executed normally, different blocks of the nonvolatile memory. Retrying to write data to the If it is determined that the writing of the data has not been normally performed by the determination step, the error correction step of reading data from the block to which the data was written at the time of the retry and performing error correction; The management table is updated with the block used for retry as a bad block only when it is determined that the error is exceeded, and when the error is determined to be within the correction capability, the error is terminated without determining the block as a bad block. This is a data writing method.
[0013]
As a result, the frequency of occurrence of error termination is reduced by performing retry processing when data writing is abnormal, and a bad block is set only when error correction is impossible even if data writing is abnormal after retry processing. This has the effect of preventing an increase in bad blocks.
[0014]
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0015]
(Embodiment)
FIG. 1 is a block diagram showing a configuration of a memory card according to an embodiment of the present invention.
[0016]
In FIG. 1, a memory card 1 is used by being mounted on an electronic device (not shown) such as a digital still camera or a personal computer. The host card 2, a controller 3, a RAM 4, An instruction ROM 5, a buffer memory 6, and a rewritable nonvolatile memory such as a flash memory 7 are provided.
[0017]
The host interface 2 is a block that interfaces with an electronic device main body (hereinafter, referred to as a host) on which the memory card 1 is mounted, and transmits and receives data to be read from and written to the flash memory 7 and commands and clocks for various controls. .
[0018]
The controller 3 is a block that controls the entire memory card 1 and has a microprocessor (hereinafter, referred to as an MPU) 31. The MPU 31 controls the RAM 4, the flash memory 7, and the buffer memory 6 according to a control program stored in the instruction ROM 5.
[0019]
The instruction ROM 5 is a read-only nonvolatile memory, and stores a control program of the MPU 31.
[0020]
The RAM 4 is a volatile memory, and includes a work RAM 41 that is a work area of the controller 3 and a block management table 42. The block management table 42 is a table for managing the use state of the flash memory 7, and an example is shown in FIG.
[0021]
Generally, reading from and writing to a physical block (hereinafter, referred to as a physical block) of the flash memory 7 is performed by designating the address of a logical block (hereinafter, referred to as a logical block) provided by a host. One role of the block management table 42 is to convert the address (logical address) of this logical block into the address (physical block) of a physical block.
[0022]
FIG. 2 shows the correspondence between a physical address and a logical address, which are block addresses, and whether or not the physical address (the block specified by the address) is used.
[0023]
For example, as shown in FIG. 2, whether or not a physical block is used and the number of a logical address assigned to each physical block are described. If the logical address is not written in the column indicating whether or not the physical block is used (ie, the physical address is “6” in the example of FIG. 2), the physical block is a bad block. , And is not selected as the write address.
[0024]
The buffer memory 6 is a memory for temporarily holding data transferred from the host. The flash memory 7 is a memory for writing data transferred from the host.
[0025]
The data writing process of the memory card 1 configured as described above will be described with reference to the flowchart shown in FIG. The flowchart shown in FIG. 3 is executed by the controller 3.
[0026]
First, as a step 1, when the controller 3 receives a write command from the host via the host interface 2, the controller 3 temporarily holds data sent from the host in the buffer memory 6, and stores the data in the flash memory 7 in block units. Transfer data to
[0027]
Next, in step 2, the MPU 31 refers to the block management data written in the block management table 42, and writes the physical address where the writable physical address, that is, whether the management data (use or not) is “1” (unused physical address). (Physical address).
[0028]
Next, as step 3, writing is performed on the physical address searched in step 2. This step 3 corresponds to data writing means or data writing step.
[0029]
Next, in step 4, a status read command is issued to the flash memory 7, and it is determined from the response from the flash memory 7 whether the processing status of the data writing has been normally performed. Step 4 corresponds to a determination unit or a determination step.
[0030]
If it is determined in step 4 that the data writing has been performed normally, in step 5 the corresponding management data on the block management table 42 is rewritten from “1” to “0” and the logical block number (logical Address) is written, and the writing ends normally (step 6). Steps 5 and 6 correspond to means for ending normally or steps for ending normally.
[0031]
On the other hand, if it is determined in step 4 that the data writing was not performed normally, it is determined in step 7 whether or not the writing process was a retry process. If it is not a retry process, the process proceeds to step 2 to execute the first retry process. Steps 7, 2, and 3 in which the process proceeds to step 2 and steps 2 and 3 are executed again correspond to a retry means or a retry step.
[0032]
If it is determined in step 7 that the process is a retry process, the process proceeds to step 8 where data is read from the physical block to which the data was written in the retry process, and data error detection is performed. Then, it is determined whether or not the data error can be corrected. If it is determined that the data can be corrected, the error is terminated without updating the block management table 42 (step 10). Step 8 corresponds to an error correction means or an error correction step.
[0033]
On the other hand, if it is determined in step 8 that the correction is impossible, in step 9 the corresponding management data on the block management table 42 is rewritten from "1" to "0", and the writing ends with an error (step 10). ). In this case, the writing of the logical address is not performed. Steps 9 and 10 correspond to a means for terminating the error or a step for terminating the error.
[0034]
As described above, according to the present embodiment, the retry processing is performed when the status read / write is abnormal in step 4, so that the frequency of ending the error in response to the write instruction from the host can be reduced. Further, since the bad block processing (BB processing) is performed only when the data is not read and error correction is not possible in step 8, it is possible to prevent the number of bad blocks from increasing. In addition, even if the error can be corrected in step 8, the quality of writing can be guaranteed by terminating the error.
[0035]
Although the above embodiment has been described with reference to the case where the present invention is applied to a memory card equipped with a nonvolatile memory, the present invention can be similarly applied to a nonvolatile memory incorporated in an electronic device, for example. It is needless to say that is not limited to a memory card.
[0036]
【The invention's effect】
As described above, according to the present invention, even in a non-volatile memory such as a multi-valued flash memory in which a write error is likely to occur, the retry processing is performed, thereby terminating the error in response to the write instruction from the host. The frequency can be reduced. Further, if the retry process is repeated, the number of defective blocks may increase. However, by registering only the case where the error cannot be corrected as a bad block, the increase of the bad blocks of the block is prevented, and the life of the nonvolatile memory is reduced. Can be avoided. In addition, data containing an error that can be corrected can be an error that cannot be corrected due to stress of writing of another block, etc. By preventing data containing an error from being left in the non-volatile memory at the end of writing, it is possible to provide a memory card which is resistant to writing stress of other blocks.
[0037]
As described above, by using the present invention, a memory card can be configured using a nonvolatile memory such as a multi-valued flash memory having a high frequency of write errors, and the cost of the memory card can be reduced. Become.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a memory card according to an embodiment of the present invention; FIG. 2 is a diagram showing a configuration of a block management table and an example of management data in the memory card; FIG. Flow chart showing the operation of [Description of reference numerals]
1 memory card 2 host interface 3 controller 4 RAM
5 Instruction ROM
6 Buffer memory 7 Flash memory 31 Microprocessor (MPU)
41 Work RAM
42 Block Management Table

Claims (2)

A memory card equipped with a rewritable nonvolatile memory, a management table for managing a data write state of the nonvolatile memory, and a unit for writing data to the nonvolatile memory in block units based on the management table, Determining means for determining whether or not the data writing has been normally performed; and prohibiting rewriting of the written block when the determining means determines that the data writing has been normally performed. Means for updating the management table and terminating normally, and means for retrying data writing to a different block of the nonvolatile memory when the determining means determines that the data writing has not been executed normally, The data writing by the retry is normally executed by the determination unit. If it is determined that the error has not been corrected, the error correction means for reading out data from the block in which the data was written at the time of the retry and performing error correction, and the error correction means uses the error only when it is determined that the error exceeds the correction capability. A memory card having means for updating the management table with a block that has been determined as a bad block and terminating the error without determining the block as a bad block when the error is determined to be within the correction capability. A method for writing data to a rewritable nonvolatile memory, comprising: writing data in block units to the nonvolatile memory based on a management table for managing a data writing state of the nonvolatile memory; A judgment step of judging whether or not the data has been normally executed; and, when it is judged by the judgment step that the data writing has been executed normally, the management table is configured to prohibit rewriting of the written block. Updating and terminating normally; retrying to write data to a different block of the nonvolatile memory when the determining step determines that the data writing was not performed normally; Is determined by the determination step. If it is determined that the error was not always performed, an error correction step of reading data from the block where the write was performed at the time of the retry and error correction, and a retry is performed only when the error correction step determines that the error exceeds the correction capability. And updating the management table with the block used in step (1) as a bad block and terminating the error without determining the block as an error if the error is within the correction capability.

Images