JPS59121525A - Subchannel control system - Google Patents

Abstract

PURPOSE:To prevent the reoccurrence of an error or the like, by providing a stand-by area for each subchannel on a memory and recovering the error surely without having the influence upon operations of subchannels other than the corresponding subchannel if the read error occurs. CONSTITUTION:A stand-by area for each subchannel is provided on the memory. For example, a control program is read out from a load area 5 of the control program in a memory 4 through a processor 3 of a transfer device 2 on a basis of a command from a central processing unit 1, and each step of the control program is executed. If the read error occurs when an area SCWm, where control data or the like is stored, in an address Am is read out, the detecting control program is searched on a management table 9 to check the address of a stand-by area SCW. If the stand-by area SCW in address An+1 cannot be used, the memory is so rewritten that a subchannel (m) on a correspondence table 6' between subchannels and addresses of areas SCW corresponds to the address An+1, and it is indicated that the area SCW in the address An+1 is busy on the management table 9 of areas SCW.

Description

DETAILED DESCRIPTION OF THE INVENTION (1) Technical Field of the Invention The present invention has a plurality of subchannels, and an area (on a built-in memory) provided corresponding to each subchannel to store control data, etc. The present invention relates to a technique for minimizing the influence of an error when an error occurs when reading an SCW in a transfer device (hereinafter referred to as SCW).

(2) Prior art and problems Figure 1 is a block diagram for explaining the control force formula of a conventional transfer device. It shows a transfer device that operates according to a control program loaded into its built-in memory, and 1 is a central 2 is a processing device, 2 is a transfer device, 5 is a processor, 4 is a memory, 5
is the control program load area, 6 is the subchannel and SC
W address correspondence table, 7 is SCW area, 81 to 8
n indicates a subchannel corresponding section. The SCW area stores sCW corresponding to each subchannel.

In FIG. 1, the transfer device 2 operates by automatically executing each step of a control program read from a control program load area 5 by a processor 5. The control program scans each subchannel corresponding section 81 to 8n and performs processing corresponding to the subchannel as necessary. At that time, it writes control data and transfer data to the SCW for reading. Access is made.

In such a transfer device 2, the following method is conventionally adopted as control when a read error (hereinafter referred to as read error) is detected during access to the SCW.

As an example, when a read error is detected when accessing the SCW, the processing of the processor 3 is stopped, and the central processing unit 1 etc. is requested to perform recovery processing, which reloads the control program. There is a method to clear scw4. This is the same 7j process that is performed when a read error occurs when accessing an area other than the SCW (for example, a control program area), and the control is simple. All the input/output devices, etc. affected by the system are affected, and the control program has to be reloaded, so it takes a long time to recover.

Another example of the conventional method is a method in which when a read error is detected in accessing the SCW, the processing of the processor 5 is not stopped, but all SCW areas are cleared. In this case as well, there was a drawback that all input/output devices connected to each subchannel and in operation were affected.

Furthermore, as another example of the conventional method, when a read error is detected in accessing the SCW, a method is adopted in which only the SCW area where the read error was detected is cleared without stopping the processing of the processor 5. There is. this·
In this method, only the input/output device connected to the subchannel corresponding to the SCW that caused the read error is affected, so only that input/output device needs to be restored, but access to the SCW that occurred first If the read error at the time was a latent failure of the hardware, there was a possibility that the read error would occur again, so there was a drawback that it could not be a fundamental solution ( This drawback is also present in the above two cases along with other drawbacks).

(3) Purpose of the Invention In view of the above-mentioned conventional drawbacks, the present invention provides, when a read error is detected when accessing an SCW, without affecting the operation of subchannels other than the subchannel corresponding to the SCW. -! The purpose is to provide a method to ensure the recovery of affected sub-channels and prevent read errors from occurring again.

(4) Structure and object of the invention According to the present invention, as described in the claims, an information processing system is provided between a main body system device and a peripheral device, and controls all data transfer between them. In a transfer device that has a plurality of subchannels and has a specific area corresponding to each subchannel on its built-in memory, a reserve area that can be used as an area corresponding to the subchannel on the memory is prepared in advance. An area is provided, and when an error is detected in reading the TK information from the area corresponding to the currently used subchannel on the memory, the area corresponding to the subchannel is switched to the area provided as a reserve. υ is achieved by the subchannel control force equation.

5) Embodiment of the invention FIG. 2 is a diagram showing the contents of the memory of a transfer device according to an embodiment of the invention, in which (a) is before SCW switching, (b)
) indicates after SCW switching, 4', 4'' is memory, 5' is control program load area, 6', 6'' is a correspondence table between subchannels and SCW addresses, 7' is SCW area, 9 is a backup SCW management table, 10.1
0' is a reserve SCW area.

In FIG. 2(a), in the subchannel and SCW address correspondence table 6', subchannel /I/1 is assigned to the SCW at address A1, and subchannel m is assigned to the SC at address Am.
SC'W1 is associated with address A1 of SCW area 7', and SCW is associated with address Am.
m exists.

Now, assume that a read error has occurred while reading SCWm'e at address Am. The control program of the transfer device that detected this is the backup SCW management table 9? Search for available (unused) spare SCs
Check W's address. And @An+1 reserve S
Assuming CW was available, the control program would be
The subchannel m on the subchannel and SCW address correspondence table 6' is rewritten to correspond to address An+1, and the SCW address at address An+1 is rewritten in the backup SCW management table 9.
The CW displays that it is in use.

As a result, the contents of the memory are shown in FIG. 2(b). J
: Become a snarl. In other words, the SCW m of the An address in which the read error was detected during reading is the SC of the An+1 plan.
W(m).

Thereafter, the central processing unit 2 performs recovery processing on the data of the input/output device connected to the subchannel m.

Data transfer of input/output devices connected to other subchannels continues regardless of the occurrence of the read error and its recovery processing, and is not affected at all.

(6) Effects of the Invention According to the method of the present invention, when a read error occurs when accessing the SCW in a transfer device having a plurality of subchannels, processing of the processor is not stopped.
Since it is only necessary to immediately switch to a backup SCW and perform recovery processing only on the data of the input/output device connected to the subchannel corresponding to the SCW, the range of damage can be minimized, and in addition, This is highly effective as it allows for quick recovery.

In other words, even if a read error is caused by a hardware failure in the SCW area, the Bokuto method uses an SCW in a different area, so the read error will not occur again and the recovery process will be completed. The effect is great because the stable movement after that is particularly effective.

[Brief explanation of drawings]

FIG. 1 is a block diagram for explaining the control force formula of a conventional transfer device, and FIG. 2 is a diagram showing the entire contents of the memory of a transfer device according to an embodiment of the present invention. , (b) are diagrams showing the state after SCW switching. DESCRIPTION OF SYMBOLS 1...Central processing unit, 2...Transfer device, 5...Processor, 4.4', 4''...Memory, 5.5'...
- Control program load area, 6.6', 6''...correspondence table of subchannels and SCW addresses, 7.7'
...SCW area, 8]~8n...Subchannel corresponding section, 9...Spare SCW management table, 10.10'
...Reserve SCW area

Claims (1)

[Claims] In an information processing system, a transfer device that is located between a main system device and a peripheral device and controls the transfer of data between them, has multiple subchannels and has a built-in memory that corresponds to each subchannel. In a transfer device having a specific area, a spare area that can be used as an area corresponding to the subchannel is provided in advance on the memory, and an error occurs when reading information from the area corresponding to the current subchannel on the memory. A subchannel control force type characterized in that when detected, the area corresponding to the subchannel is switched to the area provided as a reserve.