JP2570995B2 - Disk controller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk controller, and more particularly to a disk controller having a function for improving reliability.

[0002]

2. Description of the Related Art In a conventional disk controller, an error occurring in the disk controller or an error occurring in the disk device is detected to recover or re-access the error, thereby improving reliability. . For example, Japanese Patent Application Laid-Open No. 5-62370 describes a magnetic disk control device that automatically retry the reading process when a reading error occurs.

[0003]

In this prior art,
Even if the errored access itself can be dealt with, but cannot be involved in subsequent disk accesses, even if the error occurs frequently on a specific disk controller or channel, the error will not occur. There is a problem that every time an error occurs, the only solution is to retry. Also, there is no way to know the reliability of data already written to the disk device, and it was not possible to distinguish and use only reliable data.

[0004]

In order to solve the above-mentioned problems, according to the present invention, a channel of a higher-level device and a disk drive are provided.
Between the host device and the disk device
Disk control unit that controls the transfer of data between
The write data transferred from the higher-level device.
If the identification information indicating the transfer path of the write data
When detected, add it and transfer it to the disk drive
Identification information adding means, and
Acquisition for acquiring the identification information included in the read data
Means and said identification information obtained by said obtaining means.
Determining the reliability of the read data based on the
Notification means for notifying the position device.

Further , another disk control device of the present invention comprises:
The identification information includes a channel number, a disk controller number,
It is characterized by including.

Further , another disk control device of the present invention comprises:
The identification information includes a writing date.

[0007] Further, another disk control device of the present invention comprises:
The disk controller detects an error in the data.
Fault detecting means, and the fault detecting means detects the fault.
Identification information in which the identification information of the detected data is stored
Storage means, wherein the notifying means is the obtaining means
And the identification information obtained by
The read data is compared with the stored identification information.
It is characterized by determining reliability.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, an embodiment of the disk control device of the present invention will be described in detail with reference to the drawings.

Referring to FIG. 1, one end of each of disk controllers 200 and 300 according to an embodiment of the present invention is connected to channels 110 to 140 of higher-level device 100 by signal lines 1 respectively.
11 through 141. The other ends of the disk control devices 200 and 300 are connected to the disk device 4.
00 is connected. The host device 100 is a channel 110
An access request to the disk device is issued to any of the channels 140 through 140. Thus, the disk device 400 is accessed via the disk control device connected to the specified channel. If this access is a read from the disk device 400, the read data is transferred to the host device via the same disk control device and channel as the destination.

Referring to FIG. 3, with respect to a data format exchanged between the disk control device and another device, FIG. 3A shows a data format from the host device 100, and FIG. FIG. 3C shows a data format to the host device 100. The data section in FIG. 3A includes data to be written to the disk device. This data portion is transferred from the host device 100 in response to a request from the processor 220. The identification information in FIG. 3B includes information for matching check described later.
The match information in FIG. 3C includes a result of a match check described later.

Referring to FIG. 2, a disk controller 200 according to an embodiment of the present invention includes a host switch 210.
, Processor 220 and channel number setting register 2
21, a disk controller number setting register 222,
Write date setting register 223 and identification information adding circuit 23
0, a channel number holding register 231, a disk controller number holding register 232, a write date holding register 233, a lower switch 240, an identification information separating circuit 250, and an identification information holding register 260. .

The upper switch 210 is connected to the signal line 111 or 1 from the channel 110 or 120 of the upper device 100.
When a command to the disk device 400 is received via 21, the command is passed to the processor 220. This command includes instruction information from the host device 100 to the disk control device, and is interpreted by the processor 220. If the processor 220 identifies that this command is a write request to the disk device 400, the processor 220
Makes a request to the host device 100. Upper switch 21
0 indicates that when the write data of FIG.
Pass to. If the read request is from the disk device 400, the host device 210 receives the read data from the identification information separating circuit 250 when the read data from the disk device returns, and
The information for matching check from 0 is received, and FIG.
Is returned to the host device 100 in the format described above. At this time, the upper switch 210 waits for the read data in a state where the read request has been received, so that the data is returned to the signal line where the read request was made, and eventually the data is sent to the channel where the read request is issued. Will be returned.

The identification information adding circuit 230 is connected to the host device 21.
0 to the disk device 400,
Identification information including a channel number, a disk controller number, and a disk controller number is added. That is, the channel number held in the channel number holding register 231, the disk controller number held in the disk controller number holding register 232, and the writing date held in the writing date holding register 233 are added as identification information. The channel number holding register 231 is
Each time there is an access request from a channel having 0 to the upper switch 210, the channel number is set by a processor 220 that has recognized the channel number through a path (not shown). The disk controller number holding register 232 is set with the disk controller number at the time of system startup. Further, in the write date holding register 233, the date of the write is set by a processor 220 having a date management mechanism through a path (not shown).

The lower switch 240 transfers the data and the identification information from the identification information adding circuit 230 to the disk device 400 via the signal line 201 as the data format of FIG. Also, the lower switch 240
3 from the disk device 400 via the signal line 201.
The data read as the data format of -b) is passed to the identification information separating circuit 250.

The identification information separating circuit 250 passes the data portion of the data format of FIG. 3B received from the lower switch 240 to the upper switch 210 and the identification information to the identification information holding register 260. The identification information held in the identification information holding register 260 is read by the processor 220.

The processor 220 reads the contents of the identification information read from the identification information holding register 260 and the contents of the register 22.
It is determined whether or not the following matching condition has occurred by comparing with the values of 1 to 223. That is, the channel number of the identification information and the channel number stored in the channel number setting register 221, the disk control device number of the identification information and the disk control device number stored in the disk control device number setting register 222, and the writing of the identification information The write date and the write date stored in the write date setting register 223 are compared. If the channel number or the disk control unit number matches, or the writing date of the identification information indicates a date after the writing date held in the writing date setting register 223, the fact is described. The identification information separation circuit 2
The data is transferred to the higher-level device 100 via the higher-level switch 210 together with the data from the node 50. At this time, the upper switch 21
Since 0 is the state at the time of the access request from the higher-level device 100, desired data is transferred to the channel for which the access request has been made.

The registers 221 to 223 are stored in the host device 10.
The value is set by the processor 220 designated from 0. That is, the channel number specified by the higher-level device 100 is stored in the channel number setting register 221, the disk control device number specified by the higher-level device 100 is stored in the disk control device number setting register 222, and the write date specified by the higher-level device 100 is written. Is set in the write date setting register 223. When detecting that an error has occurred in a certain disk control device, the higher-level device 100 stores the channel number that issued the access that caused the error, the disk control device number that caused the error, and the date that the error occurred in the disk controller. To the disk control unit. Thereby, when the data is read out thereafter, whether the data has passed through the channel or the disk control device in which the error has occurred,
Alternatively, it is possible to determine whether the data is data written before an error occurs.

Next, the operation of one embodiment of the electronic disk control device will be described.

FIG. 4 is a flow chart of the operation of the error detection processing of this embodiment, which comprises three steps 271 to 273. The disk control device 200 has an error detection mechanism such as a parity check inside the device, and constantly monitors whether an error has occurred. Then, when it is detected in step 271 that an error has occurred,
The host device 100 is notified that an error has occurred (step 272). Host device 100 receiving this error notification
Indicates the channel number, the disk controller number, and the date to all the disk devices connected via each channel as described above. In each of the disk controllers receiving this instruction, the channel number setting register 221
The disk controller number setting register 222 and the writing date setting register 223 are set (step 27).
3).

FIG. 5 is a flow chart of the operation of the coincidence detection processing of this embodiment, which comprises five steps of steps 281 to 285. Upon receiving the read request from the host device 100, the disk control device 200 reads the data from the disk device 400, and passes the data portion to the host switch and the identification information to the identification information holding register 260 as described above (step 281). The processor 220 reads the contents of the identification information read from the identification information holding register 260 and the register 2
It is determined whether or not the above-mentioned matching condition has occurred by comparing the values 21 to 223 (step 282). If the matching condition has been generated, the fact that "matched" is obtained as the matching information (step 283), and if the matching condition has not been generated, the fact that "unmatched" is obtained as the matching information (step 284), and transferred to the higher-level device together with the data. (Step 285).

By receiving the coincidence information, the host device 100 can easily and accurately determine whether the data received at the same time is reliable. That is, it is possible to determine whether or not the data may have been corrupted due to passing through the channel or disk control device in which the error occurred, or data written after the occurrence of the error. become able to.

[0022]

As described above, in the disk control device according to the present invention, the channel number and the disk control device number used when storing data in the disk device and the storage date are added and stored. Depending on whether the data has passed through the channel or disk controller where the error occurred, or was stored after the error occurred,
Can be easily known, and only reliable data can be distinguished and used.

[Brief description of the drawings]

FIG. 1 is an overall view of a disk system using an embodiment of a disk control device of the present invention.

FIG. 2 is a block diagram showing one embodiment of a disk control device of the present invention.

FIG. 3 is a data format of an interface with an external device in an embodiment of the disk control device of the present invention.

FIG. 4 is a flowchart of an operation of an error detection process in an embodiment of the disk control device of the present invention.

FIG. 5 is a flowchart of an operation of a match detection process in an embodiment of the disk control device of the present invention.

[Explanation of symbols]

 100 upper device 110, 120, 130, 140 channel 200, 300 disk controller 210 upper switch 220 processor 221 channel number setting register 222 disk controller number setting register 223 write date setting register 230 identification adding circuit 231 channel number setting register 232 Disk control device number setting register 233 Write date setting register 240 Lower switch 250 Identification information separation circuit 260 Identification information holding register 400 Disk device

Claims (4)

(57) [Claims] 1. The communication between a channel of a host device and a disk device.
Provided between the host device and the disk device
In a disk controller that controls data transfer
The write data transferred from the host device.
An error was detected in the identification information indicating the data transfer path.
Identification information to be added to the disk device and transferred to the disk device
Adding means , included in the read data transferred from the disk device
Acquisition means for acquiring the identification information to be obtained , based on the identification information acquired by the acquisition means.
Judge the reliability of the read data and send it to the host device.
And a notifying means for notifying.
Control device. 2. The apparatus according to claim 1, wherein the identification information includes a channel number and
2. The control device number according to claim 1, wherein
Disk unit. 3. The identification information includes a writing date.
2. The disk drive according to claim 1, wherein: 4. The disk control device according to claim 1 , wherein
Fault detection means for detecting an error, and data of the fault detected by the fault detection means .
An identification information storage unit in which the identification information is stored.
The notifying means includes identification information obtained by the obtaining means.
And the identification information stored in the identification information storage means
And determining the reliability of the read data.
The disk device according to claim 1, wherein