JPS59211158A - Allotting system of alternate block - Google Patents

Abstract

PURPOSE:To reduce the load of software by saving the data recorded to the data part of a defective block and then writing said data to the data part of an alternate block. CONSTITUTION:For instance, a block A is defective and an alternate block B is not allotted yet. Under such conditions, an instruction indicating the defective block A is first delivered to a magnetic disk device 1 by a format defective subcommand. The device 1 reads the data part of the block A into a buffer 5, and the block B is searched an alternate area having the smallest seek time. Then an ID part of the block A is written with a defective flag and then read out. If an error is detected, the ID part is written again with a shift of several tens to several hundreds of bytes. In case the reading is through with the ID part eith no error detected, the ID part of the block B is written with an alternate block flag and an allotted flag. Then the data read by the buffer 5 is written to the data part of the block B.

Description

Detailed Description of the Invention (a) Technical Field of the Invention The present invention relates to a fixed-length block magnetic disk device, and in particular, when the data read from the magnetic disk device by a magnetic disk control device can be guaranteed, software intervention is required. The present invention relates to a replacement block allocation method for moving data from a defective block to a replacement block without any problem.

(b) Prior Art and Problems Due to manufacturing problems, magnetic disk media used in magnetic disk drives have defects on the surface of the medium, which may cause errors when reading recorded data. As shown in Figure 1, the sector format in a fixed-length block magnetic disk device is such that each sector that divides each trunk is provided with an ID section for recording index information and a DATA section for recording data. ing. I due to media defects
If there is an error in the information in the D part or if a correctable error occurs in the DATA part, the data can be guaranteed. A spare area is provided in the area, and the data is transferred to the spare block of the area.

In a conventional fixed-length block magnetic disk device, replacement blocks are allocated using a format defective subcommand of a software locate command. When the subcommand is given to the magnetic disk control device from the host device, it operates as described below.

■Search for an unallocated replacement block from the replacement area with the shortest seek (operation to position the magnetic head at the target track) time.

(2) After finding an unallocated block in the spare area, write the ID part of the defective block with a defect flag attached.

Read the ID part of the 0th defective block. If an error occurs due to the continuous writing, try writing again at a position shifted by several tens to hundreds of bytes.

- Try reading the ID part of the bad block again. If there is an error, the ID section is written again with a shift of several tens to hundreds of bytes.

(2) If the ID section is further read and an error still occurs, an error report is sent to the host device.

If the reading of the ID part of the defective block is successful without any error in the operations from 2 to 2, the ID part of the replacement block is written with the replacement block flag and the allocated flag attached.

This completes the replacement block allocation process, but writing the data written in the DATA section of the defective block to the DATA section of the replacement block is done by software, which has the drawback of increasing the load on the software.

(C) Object of the Invention The object of the present invention is to eliminate the above-mentioned drawbacks, and when allocating a replacement block, the magnetic disk control device saves the data of the defective block in a buffer without relying on software, and allocates the replacement block. After processing is completed, D of the replacement block
The object of the present invention is to provide a replacement block allocation method that reduces software load by writing data saved in the buffer into an ATA section.

(d) Structure of the Invention The structure of the present invention is that in a magnetic disk control device that controls a fixed-length block magnetic disk device, when a replacement block for a bad block is allocated, data recorded in the data section of the bad block is The vagina data is saved in a buffer, and then the vagina data is written to the data section of the replacement block.

(e) Embodiment of the Invention FIG. 2 shows an example of the configuration of a magnetic disk subsystem illustrating an embodiment of the invention. The magnetic disk control device 1 connects the magnetic disk control adapter 2 according to a command from the host device.
It controls the magnetic disk device 3 or 4 via the magnetic disk device 3.4, and writes/reads data to/from the magnetic disk of the magnetic disk device 3.4. FIG. 3 shows an example of a track for a cylinder.

For example, assume that the block indicated by A is defective and an error occurs in the ID section or the DATA section. Block B of replacement area
If it is unallocated, the operation will be as explained below.

(1) An instruction to mark block A as defective by a format defective subcommand is issued from the host device to the magnetic disk control device 1.

(2) The magnetic disk control device 1 is a block AODATA
part into buffer 5. At this time, if a correctable error occurs, it is corrected and read, but if it is uncorrectable, it is processed in the same manner as before and the present invention is not applied.

(3) Search for an unallocated block in the spare area with the shortest seek time. In this example, it is B.

(4) Write the defective block AOID section with a defective flag attached.

(5) Read the defective block AOID section. If an error occurs, the ID section is written again at a position shifted by several tens to hundreds of bytes.

(6) Try reading the ID section again. If there is an error, the ID section is written with a further shift of several tens to hundreds of bytes.

(7) Furthermore, try reading out the ID section. If an error still occurs, the error is reported to the host device.

(81 (If there is no error in the operations 41 to (7)) (reading of the ID section is completed, write the replacement block BOID section with the replacement block flag and allocated flag attached.

(9) Write the data read into the buffer 5 in (2) into the replacement block BODATA section.

(f) As described in detail, the present invention uses software to enable the magnetic disk control device to independently execute the data recorded in the data section after the replacement block has been allocated, without any software intervention. It is possible to reduce the load.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a sector formant, FIG. 2 is a diagram showing an example of the configuration of a magnetic disk subsystem explaining an embodiment of the present invention, and FIG. 3 is a diagram showing an example of a track of a certain cylinder. It is. 1 is a magnetic disk control device, 2 is a magnetic disk control adapter, 3 and 4 are magnetic disk devices, and 5 is a buffer. Kaya 1 Diagram 2 Diagram 3 Diagram

Claims (1)

[Claims] In a magnetic disk control device that controls a fixed-length block magnetic disk device, when allocating a replacement block for a bad block, the data recorded in the data section of the bad block is saved in a buffer, and then the vaginal data is replaced. A replacement block allocation method characterized by writing to the data portion of a block.