JP3457602B2 - Disk unit - Google Patents

Description

Detailed Description of the Invention

[0001]

The present invention has a function of providing useful information for cause analysis when a failure occurs .
Disc device

[0002]

2. Description of the Related Art Magnetic disk devices are widely used as external storage devices for computers. In a magnetic disk device, data recording (or reading) is performed by a magnetic disk seeking operation in a radial direction of a medium with respect to a rotating disk-shaped magnetic disk. Since such a magnetic disk device has a mechanical operation part, it has a higher failure rate than, for example, a semiconductor device. Therefore, for example, it is necessary for the user to take a countermeasure against artificial failure such that the user regularly backs up the data. In consideration of such a situation, a magnetic disk device having a function of diagnosing a failure by the device itself has been commercialized. According to the self-diagnosis function of failure in the magnetic disk device, RAW error rate (raw error rate), CSS (Contact Start)
Stop) times, start retry, seek performance, etc.
The parameters indicating the factors leading to the failure of the device are regularly checked, and it is diagnosed whether or not there is a dangerous state in which the occurrence of the failure can be predicted. When a dangerous state is diagnosed, a warning is issued to the host system in advance.

As a more advanced self-diagnosis function, SMA
RT (Failure Prediction Mechanism) is known. SMART is A
Supported by TA (AT Attachment) standard. The ATA standard is an IDE interface that is a practical standard for connecting a personal computer and a magnetic disk device, and is a standard standardized by the American National Standards Institute (ANSI).

One of the functions of SMART is a function of acquiring an event log. This function is a log (history) of various events specified in the magnetic disk device.
Is recorded on a medium, and the recorded event is used for failure diagnosis and the like.

As a more specific structure, 255 blocks are allocated as a log recording area. One block consists of 16 sectors at the maximum. A out of 255 blocks
128 blocks as an area where the TA standard sets the specifications,
32 blocks are allocated for the host vendor and 96 blocks for the device vendor.

At present, two blocks defined by ATA are used, and by using these two blocks, a command status log up to five commands back from the time of error occurrence and the error status is recorded. An error log function for recording and a function for recording an execution result log of a self-diagnosis (self-test) function for diagnosing the processing function of the magnetic disk device are realized. The log information recorded by these functions can be read by a subcommand implemented by the SMART mechanism. The host system can also write and read unique information in the log recording area of the medium.

Such a conventional example has the following problems. That is, when a failure such as an operation abnormality occurs on the host system side, the self-diagnosis function provided by the conventional magnetic disk device is not always sufficient to obtain the information necessary for the cause analysis. There is a point. For example, this may be the case when a failure occurs due to a malfunction of the driver software on the host system side.

[0008]

SUMMARY OF THE INVENTION An object of the present invention is to provide a disk device which can obtain useful information for cause analysis when a failure occurs.

[0009]

A disk device according to the aspect of the present invention responds to a command issued from a host.
The disc to record or play data.
Disk device that receives a frame from the host
Command register that stores the command and the command
Status register that sets a flag that indicates the execution status
Register means including a command received from the host
Mode is stored in the command register,
Received in the status register prior to the command
The flag is set to indicate that the command is running
If it is, the command overwrite has occurred.
Creator to create command overwrite information indicating
And a memory for storing the command overwrite information.
And means .

[0010]

[0011]

[0012]

[0013]

[0014]

[0015]

[0016]

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) FIG. 1 is a block diagram showing the schematic arrangement of a magnetic disk apparatus according to the first embodiment of the present invention. As shown in the figure, the magnetic disk device of this embodiment has a nonvolatile recording / reproducing medium 1 (medium), a read / write circuit 2, a disk data controller 3, a buffer RAM 4, a CPU 5, a host interface 6, a ROM and a RAM 7. , A VCM (voice coil motor) and a spindle motor driver 8, and a disk head 10.

The disk head 10 is attached to a head moving mechanism called a carriage. The disk head 10 can be moved in the radial direction of the medium 1 by the operation of the carriage. Disk head 10
Is controlled by a servo processing system (head positioning control mechanism) on a designated target cylinder (target track). Normally, read / write data in sector units
The light is written.

The carriage is driven by the VCM. The VCM is driven by the driver 8. The spindle motor is also driven by the driver 8. Driver 8
Is a driver IC (integrated circuit) that constitutes a motor driver and a VCM driver (referred to as a double driver)
Is.

The read / write circuit 2 receives the read signal from the disk head 10 amplified by the head amplifier and performs signal processing necessary for data reproducing operation. The read / write circuit 2 also performs signal processing necessary for data recording operation, and supplies a write current according to write data to the disk head 10 via a head amplifier.

Further, the read / write circuit 2 executes a normal user data recording / reproducing process and a servo data reproducing process required for a servo process for head positioning control. The servo data includes a cylinder address indicating the current position of the disk head 10 and burst data (burst signal) indicating a position error within the cylinder. A servo circuit (not shown) extracts and holds the cylinder address from the data pulse output from the read / write circuit 2. Further, the servo circuit outputs the sample timing signal for extracting the burst signal or extracts the sector pulse.

The CPU 5 constitutes a servo processing system for executing head positioning control together with the servo circuit. The CPU 5 executes various drive controls of the magnetic disk device including read / write data transfer control, in addition to head positioning control. In such a case, the CPU 5 has a ROM
Various drive controls are executed based on the control program stored in 7. The ROM 7 is connected to the CPU 5 via an external bus. The ROM 5 stores firmware for failure diagnosis of the magnetic disk device.
The ROM 7 also stores conditions such as fluctuations in the output level of data, the number of times read processing has been repeated (the number of retries), and the fact that data has been correctly read by an error correction function such as ECC. Further, the firmware of the ROM 7 has a function of collecting command issue history information according to the feature of the present invention. The details will be described later.

The CPU 5 comprises a one-chip microprocessor, and has an A / D converter and a D as an input / output unit.
Built-in / A converter. Further, it has an internal read / write memory called an instruction RAM. In the embodiment, the CPU 5 uses an A / D converter to convert the supplied burst signal into digital data. Also, CP
U5 uses a D / A converter to convert the control amount required for servo processing (control amount required for head positioning control etc.) into an analog signal and output it to the driver 8.

The disk data controller 3 is a controller that constitutes an interface between the magnetic disk device and the host system and mainly transfers read / write data. This disk data controller 3
Stores the read data read from the medium 1 in sector units and the write data for writing in the medium 1 in the buffer RAM (RAM) 4 temporarily. The disk data controller 3 and the CPU 5 constitute a system that controls data transfer. The disk data controller 3 is also provided with an error detection circuit. This error detection circuit has a function of detecting a data error based on the data read from the medium 1 and a function of recovering an error in a predetermined narrow area, that is, an ECC (Error Correction Code) function. There is.

The disk data controller 3 also has a host (not shown) via the host interface 6.
It has a register group 12 used when processing a command sent from. The register group 12 is also called a task file, and includes a status register 121, a command register 122, and an address register 123, as shown in FIG. The register group 12 has registers other than these, but the description thereof is omitted here. The disk data controller 3 also has a command sequence information storage unit 14. The storage unit 14 is specially prepared in a predetermined area in the storage unit supported by the log function of the SMART (fault prediction) mechanism described above. In addition,
The storage unit 14 may be provided in the buffer RAM 4.

The status register 121 is a register indicating the execution state of the command written from the host to the command register 122. In the disk data controller 3 (on the magnetic disk device side), the register 121 is set to a busy state while the command is being executed, and when the execution is completed, the register 121 is ready.
The state is set. These are expressed inside the firmware as set / reset of command execution flags. The host (not shown) is a status register 121.
The command execution status can be detected by referring to the command. Therefore, the command is sent at an appropriate timing during normal operation. However, if some trouble occurs due to a software defect on the host side, the status register 121 is busy. Despite the status, the subsequent command is supplied from the host, and the command register 122
May be written in. This is called command overwrite.

The magnetic disk device of this embodiment comprises means for detecting command overwrite in the disk data controller 3 and recording the detection result in the command issue history information. This is realized as an additional function of the firmware described above.

The operation at the time of command execution by the magnetic disk device of the present embodiment configured as described above will be described with reference to the flowchart of FIG. First, in step S1, a command is supplied to the disk data controller 3 via the host interface 6. This command is written in the command register 122 of the register group 12. At this time, the content of the task file (value of each register) corresponding to the command and the lifetime stamp are temporarily stored in the buffer RAM 4 as command information. Here, the lifetime stamp refers to the value of the elapsed time since the magnetic disk device was powered on.

The command information is also used for other error log functions of the SMART mechanism. The command information storage area constitutes a ring buffer of finite length, and when the storage data amount becomes full, the latest information is overwritten with the oldest data.

Here, it is determined whether or not command overwrite has occurred based on the command execution flag (step S2). That is, even if the command execution flag is set, the step S2
The occurrence of command overwrite is detected when is executed. At this time, the subsequent command supplied from the host is written in the command register 122. Note that the detection method here is merely an example, and can be replaced by another detection method.

When command overwrite is detected, the detection result is stored in the storage unit 14 as command overwrite information. The storage unit 14 constitutes a finite-length ring buffer like the command information,
The command overwrite information of multiple cases is stored as a log. When the amount of stored data becomes full, the latest information is overwritten with the oldest data. The top area of the storage unit 14 is
Has a log index. The log index is
It indicates the registration order of the command overwrite information, such as the number of the latest one logged. Following the log index, command overwrite information is stored.

The command overwrite information has a data size of, for example, 512 bytes per item, and the contents thereof include the value of the task file of the command issued earlier and the execution of which has not been completed, and the value of the lifetime stamp. , The value of the task file of the command sent this time,
It is the value of the lifetime stamp and the value of the issuance time interval (ms) of these two commands. The data structure (format) of the command overwrite information is not limited to this.

Such command overwrite information is recorded as log data for each detection, but the command overwrite newly detected next time corresponds to the subsequent area in the storage unit 14. Command overwrite information is written. When the storage capacity of the storage unit 14 is full, the latest data is overwritten on the oldest data.

On the other hand, if the command overwrite has not occurred, the command execution flag is set corresponding to the command this time (step S4). Then, the command is executed (step S5). When the execution is completed, the command execution flag is reset (step S6).

According to the magnetic disk device of this embodiment as described above, the command overwrite information is stored in the storage unit 14.
Remember. Therefore, by reading the stored command overwrite information, it becomes possible to know that the subsequent command is supplied from the host before the completion of execution of the preceding command issued from the host. Therefore, it is possible to obtain useful information for cause analysis when a failure occurs.

(Second Embodiment) FIG. 4 is a block diagram showing a schematic configuration of a magnetic disk device according to a second embodiment of the present invention. In this device, the same parts as those in the first embodiment are designated by the same reference numerals.

The magnetic disk drive according to the second embodiment is
This is different from the first embodiment in that command information including command overwrite information (collectively referred to as command information) is recorded in a nonvolatile recording medium. Particularly, here, a medium area 1 specially provided in the recording / reproducing medium (media) 1.
The command information is recorded in 1. Further, in this embodiment, the medium area 11 of the medium 1 is responded to in response to a predetermined request (command).
Command information recorded in the host interface 6
It is a configuration for transferring to the host via.

In the present embodiment, information corresponding to the above-mentioned command overwrite information is stored in the storage unit 14 regardless of the occurrence of command overwrite, and
The information is sent to the medium area 11 of the medium 1 at an appropriate timing.
Write to. That is, the log writing operation is performed excluding the time of receiving a command so that the time required for writing data to the medium 1 via the read / write circuit 2 does not affect the operation of the entire magnetic disk device. Be seen. A ring buffer is also configured for the medium area 11 of the medium 1, and when the recorded data in the area is full, the oldest data is overwritten with the latest data.

When the host receives the transfer of the log information recorded in the medium area 11 of the medium 1 of the magnetic disk device, both the command information and the command overwrite information are displayed, and only the command information is displayed. In some cases, only (that is, the command issue history) is displayed.

Therefore, when the host requests the magnetic disk device to transfer the log information, a means for instructing whether or not to display the command overwrite information is provided. Specifically, a special bit is prepared in the Feature register or the like related to the transfer request.

In such a case, the magnetic disk device of this embodiment transfers both the command information and the command overwrite information to the host via the host interface 6 when the command overwrite information is instructed to be displayed. On the other hand, when the display of the command overwrite information is not instructed, only the command information is transferred to the host via the host interface 6. The host can display the sent information in its original format.

According to the magnetic disk device of the second embodiment as described above, the same effect as that of the first embodiment can be obtained, and the command overwrite information is recorded in the nonvolatile storage medium (medium 1). As a result, the information can be obtained permanently. Also, regarding transfer of the record information to the host, it is possible to specify whether to display the command overwrite information,
A flexible system configuration can be obtained.

The present invention is not limited to the above-described embodiment, but can be implemented with various modifications. For example, the present invention can be applied not only to the magnetic disk device but also to other information recording devices.

[0045]

As described above, according to the present invention,
A disk device that can obtain useful information for cause analysis when a failure occurs can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a magnetic disk device according to a first embodiment of the present invention.

FIG. 2 is a diagram showing a detailed configuration of a register group (task file).

FIG. 3 is a flowchart showing an operation when a command is executed in the magnetic disk device.

FIG. 4 is a block diagram showing a schematic configuration of a magnetic disk device according to a second embodiment of the present invention.

[Explanation of symbols]

1 ... Recording / reproducing medium 1 (media) 2 ... Read / write circuit 3 ... Disk data controller 4 ... Buffer RAM 5 ... CPU 6 ... Host interface 7 ... ROM and RAM 8 ... VCM (voice coil motor) and spindle motor driver 10 ... disk head 12 ... register group 14 ... command sequence information storage unit

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G06F 3/06-3/08 G11B 20/10 G11B 20/18

Claims (3)

(57) [Claims] 1. A command issued by a host
The disc to record or play data.
A disk device that stores a command received from the host.
Register and a flag indicating the execution status of the command.
Register means including a status register for setting and a command received from the host to the command register.
Data in the status register
The command received prior to the command is being executed
If the flag indicating that is set,
Command over indicating that a overwrite has occurred
A disk device comprising: a creating unit that creates write information, and a storage unit that stores the command overwrite information . 2. The command and the data are exchanged with the host.
A controller for executing data transfer, the controller including the register means and the storage device.
A command step stored in the storage means, including a step.
Characterized in that the bar light information is transferred to the host.
The disk device according to claim 1, further comprising: 3. According to the command issued from the host
The disc to record or play data.
A command processing method applied to a disk device in which a command received from the host is stored in a command register.
Storing step and when the command is stored in the command register
, A flag that indicates the execution status of the command is set
Refer to the status register and precede the command.
A flag indicating that the received command is being executed is set
Command overwrite is issued,
Command overwrite information indicating that
And a step of storing the command overwrite information in the storage means.
A procedure characterized by executing a procedure having steps and
Mand processing method.