JPH07335000A - Non-volatile memory - Google Patents

Abstract

PURPOSE:To accurately and easily transmit detected result of a pad spot performed by a chip maker to an user by providing an exclusive memory block for storing all position information of the pad spot existing in the memory block. CONSTITUTION:A chip maker tests a flash memory 1, and stores the number of detected pad spots N in a region 5 for storing the number of pad spots in a specific block 2. Next, after the chip maker stores respective addresses of N pieces of pad spots, supplies the memory 1 to an user. Thereby, by reading out the specific block 2, the user can easily find a position of a pad spot in a non-volatile memory, and can take relieving measure for the pad spot.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a non-volatile memory, and more particularly to a non-volatile memory having an address space where a memory cell cannot be normally accessed.

[0002]

2. Description of the Related Art Flash memory (nonvolatile memory)
With the increasing density of memory cells, a memory cell having a large memory capacity is provided. However, the yield of memory chips is not so high, and defective memory cells are often partially produced during manufacturing. Normal data access cannot be guaranteed for such defective memory cells. In addition, normal data access may not be guaranteed even if the address decoder is defective. As described above, an address space in which normal data access cannot be performed due to a defect in the memory cell itself, a defect in the address decoder, or the like is called a bad spot.

Memory chips containing bad spots are usually judged to be defective and are not supplied to the user. However, even if the flash memory has a large capacity and the yield is low, only a small number of memory chips can be supplied to the user.
As for memory chips containing a certain number of bad spots or less, supply to users is permitted.

Since the bad spot in the memory chip cannot be used, the user needs to inspect the supplied memory chip and detect the bad spot. The detected bad spot must be replaced with another memory or the like to take measures to avoid data access to the bad spot.

The detection of bad spots is also carried out at the time of shipping inspection even at a chip maker, and only memory chips having a prescribed number of bad spots or less are supplied to the user. The shipping inspection performed by the chip maker is a test performed in a severe environment. On the other hand, since the test performed by the user is a test performed in a normal operating environment, there are cases where only a smaller number than the number of bad spots recognized by the chip manufacturer in the shipping test can be detected.

The bad spot detected by the inspection of the chip maker is not always detected by the inspection of the user. Such a bad spot can be accessed without any problems at the beginning of use of the memory chip, but becomes unstable with an increase in the number of accesses or the passage of time, and the bad spot becomes apparent and normal access cannot be performed. Sometimes.

[0007]

The user must independently detect the bad spot because there is no means for acquiring information on the bad spot detected by the shipping inspection performed by the chip maker. However, the step of detecting the bad spot is complicated, and it is difficult to obtain a detection result as reliable as the shipping inspection performed by the chip maker.

It is an object of the present invention to provide a non-volatile memory which can accurately and easily notify a user of a bad spot detection result performed by a chip maker.

[0009]

A nonvolatile memory according to the present invention includes a plurality of first memory blocks each including a plurality of memory cells, and a plurality of memory cells, and a bad spot in the first memory block. Second memory block for storing the position information of the.

[0010]

Since the memory block has a dedicated memory block for storing all the position information of the bad spots existing in the memory block, the position of the bad spot in the non-volatile memory can be known by reading the memory block.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1A is a schematic diagram showing the structure of a flash memory according to an embodiment of the present invention. The flash memory 1 has a plurality of memory blocks, and one memory block has a plurality of memory cells of 512 bytes, for example. The specific block 2 is a memory block (block number 0) located at the head of the memory blocks included in the flash memory 1. The general block 3 is a memory block remaining in the plurality of memory blocks included in the flash memory 1 except the specific block 2.

FIG. 1B is a schematic diagram showing the structure of a specific block in the flash memory 1. The chip maker inspects the flash memory 1 and prepares to register the detection result of the bad spot in the specific block 2. The inspection detects the address of each bad spot and the number N of bad spots.

In the specific block 2, first, the number N of detected bad spots is stored in the bad spot number storage area 5. Then, in the bad spot list storage area 6,
Detected N bad spots (bad spots 1 to
Store each address of the bad spot N).

As described above, after registering the bad spot information in the specific block 2, the chip maker supplies the flash memory 1 to the user. Nothing needs to be done to the general block 3.

The user reads the bad spot information from the specific block 2 located at the head of the supplied flash memory. To read the bad spot information, first, the bad spot number N is read from the bad spot number storage area 5. Next, the number of read bad spots N
Only the bad spot address (bad spot 1 to bad spot N) is read. Since the read bad spot address cannot be used, a remedy such as substitution with another memory is taken.

After that, since the bad spot information in the specific block 2 is not required, the bad spot information can be erased and the specific block 2 and the general block 3 can be used as a normal memory block. Of course, without deleting the bad spot information, the general block 3
Only one may be used as a memory block.

In the flash memory 1, when one memory block is the minimum unit of access, if there is at least one bad spot in the memory block, that memory block cannot be used. Therefore, when there are a plurality of bad spots in one memory block, only one of the bad spot addresses may be registered in the bad spot list storage area 6 in the specific block 2.

Instead of the bad spot address, the block number of the memory block containing one or more detected bad spots may be registered in the bad spot list storage area 6.

Although the case where the specific block 2 is located at the head of the flash memory 1 has been described above, it is not necessary to be at the head and any one of the last memory block and other memory blocks may be the specific block 2. .

However, in a large-capacity flash memory, when many bad spots are included, not only one memory block may be allocated as the specific block 2, but a plurality of memory blocks may be allocated as the specific block.

FIG. 2 is a schematic diagram showing a configuration example of another specific block used in place of the specific block of FIG. 1 (B). The specific block 12 has an end code indicating the end of the list of bad spot addresses, instead of the number N of bad spots. The specific block 12 has a bad spot list storage area 16 and an end code storage area 17. The bad spot list storage area 16 stores all the addresses of the detected bad spots or the numbers of the memory blocks including the bad spots (bad spot 1 to bad spot N), as in the specific block described above.

First, the user first selects the bad spot list 16
From the beginning of, bad spot 1, bad spot 2, ...
・ Read the bad spot address or block number until the end code appears in the order of. Then, when the end code is read, it is determined that the bad spot list has ended, and the bad spot is read based on the bad spot address (bad spot 1 to bad spot N) read before the end code. Take remedies for. The end code is a special code that is not used for bad spot addresses or block numbers.

In the above embodiments, it is premised that the bad spot exists in the general block 3 in the flash memory 1 and that the specific block 2 does not contain the bad spot. Therefore, the present embodiment cannot be applied when the bad spot is included in the specific block. Therefore, next, an example of a flash memory applicable to any memory block including a bad spot will be described.

FIG. 3 shows the structure of a flash memory according to another embodiment. The flash memory 21 has a memory block 22 that does not include a bad spot and a memory block 23 that includes a bad spot.

At the time of shipping inspection, the chip maker first writes predetermined test data (eg, "AA") in all the memory cells in the memory block 22. After writing the test data (“AA”) in the memory block 22, the verification is performed to detect the bad spot. When the verification is successful and there is no bad spot in the memory block 22, data erase is performed. When the data is erased, all the memory cells in the memory block 22 are filled with erase data (for example, “FF”). That is, when the memory block does not include a bad spot, the entire memory block is filled with erase data (“FF”).

Next, the case where the memory block includes a bad spot will be described. Similarly to the above, first, the test data (“AA”) is written in the memory block 23. After writing the test data (“AA”) in the memory block 23, verification is performed. If the verification fails, it means that there is a bad spot in the memory block 23. Therefore, the data block is left unerased.
As a result, as shown in the figure, the memory block 23 is basically filled with the test data (“AA”), but the bad spot portion is not correctly written and data different from the test data (for example, “AE”). In some cases, ") remains written.

After performing the bad spot detection as described above, the chip maker supplies the flash memory to the user. The user can grasp the position of the bad spot from the bad spot information registered in the supplied flash memory.

For example, it is possible to determine that a memory cell that is completely filled with erase data (“FF”) is a memory block that does not include a bad spot, and the rest is a memory block that includes a bad spot. You can judge.

Further, a memory block containing test data (“AA”) or a memory block containing data other than erase data (“FF”) can be judged to be a memory block containing a bad spot. .

For example, if the data "AA" can always be used as test data, no problem will occur.
Identifying test data that way is not always a good idea. Therefore, if the test data cannot be specified, it cannot be compared with the test data, so if the verification fails after writing the test data, a predetermined error code indicating that a bad spot exists will be output. You may make it write in the memory block. In that case, the memory block including the error code can be determined as the memory block including the bad spot.

As described above, the specific block is provided in the flash memory and the bad spot information is registered, or the information as to whether or not the bad spot exists directly in each memory block in the flash memory is registered. By doing so, the bad spot information provided by the chip maker can be accurately and easily transmitted to the user.
The user does not have to perform another test for detecting a bad spot.

The present invention has been described above with reference to the embodiments.
The present invention is not limited to these. For example, it will be apparent to those skilled in the art that various modifications, improvements, combinations, and the like can be made.

[0033]

As described above, according to the present invention,
Since the memory block has a dedicated memory block for storing the position information of the bad spot, the position of the bad spot in the nonvolatile memory can be easily known by reading the memory block, and a remedy for the bad spot can be taken.

[Brief description of drawings]

FIG. 1A is a schematic diagram showing a configuration of a flash memory according to an embodiment of the present invention, and FIG. 1B is a schematic diagram showing a configuration of a specific block in the flash memory.

FIG. 2 is a schematic diagram showing a configuration example of another specific block used in place of the specific block of FIG. 1 (B).

FIG. 3 is a schematic diagram showing a configuration of a flash memory according to another embodiment.

[Explanation of symbols]

 1, 21 Flash memory 2, 12 Specific block 3 General block 5 Bad spot number storage area 6, 16 Bad spot list storage area 17 End code storage area 22, 23 Memory block

Claims (6)

[Claims] 1. A non-volatile memory including a plurality of memory blocks each including a plurality of memory cells and including an address space (hereinafter referred to as a "bad spot") in which access to a specific memory cell cannot be normally performed. In the manufacturing of the non-volatile memory, the step of detecting the position information of the bad spot existing in the non-volatile memory and the step of registering the position information of the detected bad spot in a predetermined memory block in the non-volatile memory. Bad spot information registration method to. 2. A plurality of first memory blocks (3) each including a plurality of memory cells, and a plurality of memory cells, wherein a specific memory cell is normally accessed in the first memory block. A non-volatile memory having a second memory block (2) for storing position information of an address space that cannot be performed. 3. A non-volatile memory including a plurality of memory blocks each including a plurality of memory cells and including an address space (hereinafter referred to as a bad spot) in which access to a specific memory cell cannot be normally performed. Using the position information of the bad spot from a predetermined memory block, and determining that the memory block corresponding to the position information of the read bad spot is a memory block including the bad spot. Non-volatile memory bad spot detection method. 4. A non-volatile memory including a plurality of memory blocks each including a plurality of memory cells and including an address space (hereinafter referred to as a bad spot) in which access to a specific memory cell cannot be normally performed. In manufacturing of, a nonvolatile memory including a step of detecting a bad spot existing in the non-volatile memory and a step of storing different data in a memory block not including the bad spot and a memory block including the bad spot. Bad spot information registration method. 5. A first data is stored which does not include an address space (hereinafter referred to as a bad spot) having a plurality of memory cells, in which access to a specific memory cell cannot be normally performed. A first memory block (22) having a plurality of memory cells, a second data (error code) different from the first data to include a bad spot and to indicate the existence of the bad spot. The second memory block (2
3) A non-volatile memory having: 6. A non-volatile memory including a plurality of memory blocks each including a plurality of memory cells and including an address space (hereinafter referred to as a bad spot) in which access to a specific memory cell cannot be normally performed. When using, the steps of reading the data stored in the memory block, the step of inspecting whether the read memory block contains a predetermined error code, and the memory block containing the error code And a step of determining that a bad spot is included in the non-volatile memory.