JPS60205650A - Program loading system - Google Patents

Abstract

PURPOSE:To speed up the rising of the system by sending indication requests for memory block transfer and program loading to a central monitoring device when a part where initial program loading is not performed is accessed. CONSTITUTION:The central monitoring device 2 having received an initial program loading (IPL) request from a terminal 4 reads corresponding blocks out of a storage device 3 according to an IPL data transmission sequence table and sends them out to the terminal 4. Then, it sends an execution instruction for a program during the IPL to the terminal 4 after the blocks corresponding to a block number set in the table are sent out. The monitoring device 2 when receiving a report on the contravention to memory protection before or during the transmission of data with a specific block to the terminal 4 registers the contravention block number in the IPL transmission sequence table at the position of the last specific block number. Then, a memory address for making a restart at the block and number of memory where the contravention to the memory protection is caused by the terminal is sent out to the terminal 4.

Description

[Detailed description of the invention] (Technical field) It is.

(Prior art) Conventional systems that perform IPL via a network only load IPL data into memory until IPL is completed, so when the network traffic is large or the amount of data to be transferred is large, , ak is IP
If the sending speed of the L data is slow, it will take a very long time, and no processing will be done until the next IPL data is received.

(Object of the Invention) An object of the present invention is to provide an IPL method that eliminates the deficiencies of the conventional method and speeds up system start-up when performing IPL on a plurality of systems via a network.

(Structure of the Invention) According to the present invention, when performing IPL via a network, a device receiving IPL has means for prohibiting writing and reading of memory for each predetermined block, and Run the program I
An IPL method is obtained which is characterized in that when access is made to a part where PL is not performed, an instruction request to transfer this memory block and terminate program loading is sent to a central monitoring device connected to the network. .

(Example) Next, an example of the present invention will be described with reference to the drawings. 1st
The figure is a configuration diagram of a network to which the present invention is applied, where 1 is a data transmission path, 2 is a central monitoring device, 3 is a storage device, and 4
-6 are terminals. FIG. 2 is an explanatory diagram of the internal functions of the terminal 4 used in the present invention, where lO is an internal data bus;
is a central control unit, 12 is a ROM for performing IPL, 1
Reference numeral 3 denotes a memory protect table for prohibiting writing and reading of each block of memory, and 14 denotes an access detection circuit for blocks in which writing and reading to the memory are prohibited according to the memory protect table 15.

The central monitoring device 2 receives the request from the terminal 4, and the central monitoring device 2 receives the request from the terminal 4.
According to the IPL data sending order table in a), the block is read from the storage device 3 and sent to the terminal 4. After sending out the number of blocks set in the table, an execution command for the program being IPLed is sent to the terminal 4.

When the terminal 4 receives IPL data, it releases the memory read/write prohibition (memory protection) of the block to be transferred and then transfers it to the memory, and if one block has not all been transferred, it sets the memory protection again. When an execution command is received, the memory protection of the specified block is canceled, if any, and the program is executed from the specified position.

If IPL data is received while the IPLed program is being executed, it is transferred to the memory and executed again in the same manner as above. If memory protection is violated during execution, the central monitoring device 2 is notified of the memory block and address;
Reception of IPL data and restart instructions ◇ If it takes time to read data from the storage device 3, or if a notification of a memory protection violation is received from the terminal 4, transfer the data of the block where the memory protection was violated or restart. If it takes time to send an instruction to transfer data, the terminal 4 is instructed to transfer data to a terminal that has already transferred the same content.

Before sending the data of block number 24 in FIG. 3(a) to the terminal 4, the central monitoring device 2 is notified of the memory protection violation of block number 22. The number is re-registered before block number 24 in the IPL transmission order table for terminal 4 as shown in FIG. 3(b), and terminal 4 restarts from the memory block and address where the memory protection violation occurred. According to the invention, f
a'f3 mode If the amount of data to be transferred is large or the network traffic is large,
Even if the reading speed of the storage device storing AD-code data is slow, it is possible to start up the system quickly.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of the configuration of a network to which the present invention is applied, and FIG. 2 is a block diagram of an embodiment of a terminal device to which the present invention is applied. FIG. 3 (at(bl) is an order table of memory blocks for IPL data transmission. 1...Data transmission path, 2...Central monitoring device, 3...Storage device, 4,5.6...terminal, 10...internal data bus, 11...
・Central control unit, 12...ROM, 13...
...Memory protect table 114...Access detection circuit. ”l(y ′5 evil.

Claims (2)

[Claims] (1) When a program is loaded via a network, the device that receives the program load has means for inhibiting writing and reading of memory for each predetermined block, and the device that is loading the program before the program load is completed. ``When the program is executed and accesses a part where the program load is not performed, the program sends an instruction request to the central monitoring device connected to the network to transfer the memory block and restart the program load. Load method. (2) If the central supervisory control device cannot respond immediately to a request to send a program load restart instruction while a program is being loaded, program load data will not be sent from another device that has loaded the same program until the restart instruction is sent. Claim No. (11) characterized in that all
If it takes longer than a predetermined time to send out the program data set forth in Claim 1, the program load data is sent out once from another device that has loaded the same program. Program loading method.