JPH0474744B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method of controlling an electronic computer, and particularly to a method of controlling a management table of an operating system.

Generally, a user program in a computer usually operates in units of work called tasks. That is, a task is the smallest unit of work that is performed using resources such as a central processing unit and peripheral devices based on data accompanying a program. The operating system assigns priorities to each task, allocates resources based on efficiency, and manages task execution status for efficient scheduling.
A management table called TCB (Task Control Block) is provided.

Figure 1 shows the configuration of this TCB. TCB in operating system 10 on main memory 1
100 is TCB110 for task number=1, TCB115 for task number=2, and so on.
Each TCB up to TCB120 for task number = max
They have the same capacity and the same format and are arranged sequentially, and the leading address is also indicated by a fixed address 50 within the operating system 10. Here, the task number is a consistent number assigned to each task by the operating system to identify each task, and is given when a user program is generated as a task. next
The detailed configuration of TCB will be explained. Reference numeral 200 indicates various resource waiting pointers, and when the resources required by the task are being used by other tasks, these pointers are used to connect the task to a queue. Flags 205 indicate the status of the task (running, waiting for free resources, temporarily suspended, etc.). Among them, 255
and 260 are flags used in a system configuration consisting of a main memory and an auxiliary storage memory (hereinafter referred to as a system with DISC). 215
is an area that stores the activation factor when starting this task. 210 and 220 are used as areas for linking computers with information necessary in a system consisting of a plurality of computers (hereinafter referred to as a multi-computer system). 225 is an area for storing task attributes (task type, priority number, etc.), and 265 is an area for storing task attributes (task type, priority number, etc.).
This area is used only for systems with Reference numeral 230 is information indicating the position of the work area of the task, and stores the upper and lower limit addresses of the work area in the form of relative byte addresses from the beginning of the store area of the task. 235
indicates the start address in the main memory during the operation of this task. 240 is information necessary to control other tasks, of which 270
is information necessary only in the case of a system with a DISC, and 250 is information necessary only in the case of a multi-computer system. Each TCB is
The configuration is as described above, and the capacity of each is 37 bytes.

Now, when we look at each piece of information in this TCB, depending on the system configuration of the computer, there may be unnecessary information. 210, 220, and 250 are information required only in a multi-computer system format;
It is not necessary for other system types. Furthermore, 255, 260, 265, 270 are DISC
This information is necessary only in the case of a system with a main memory, and is not necessary in a system with only a main memory (hereinafter referred to as a main memory only system). In other words, the system configuration is a system with DISC,
Even if there are differences between a multi-computer system and a main memory only system, the TCB has the same configuration regardless of the system type. The reason for this will be explained next.

FIG. 2 shows an example of a conventional program for fetching 1 byte of data 245 from the TCB 110 of FIG. 1 into the register 10 (R10). The commands used in the program 300 will be briefly explained. LDR: Reads one word (32 bits) of data from main memory and sets it in the specified register.
LD: Read one word from main memory and set it in Acc register. M: Set the product of the contents of the Acc register and the contents of the memory in the Acc register. TTR;
Sets the contents of the Acc register to the specified register. AND: Performs a logical AND operation between the contents of the Acc register and the contents of memory, and sets the result in the Acc register. SR: Shifts the contents of the Acc register to the right by the specified bit amount.

The program 300 requires the following procedures.

Step 305; R the start address of TCB100
Set to 5 register.

Steps 310 to 320: Number of TCB bytes 37 for each task number configuring TCB100 and data 2
From case number (1) of TCB110 including 45,
Find the relative address of TCB 110 in TCB 100 and set it in IXR1 register.

Step 325: Read one word containing data 245, ie, the ninth word of TCB 110, from the main memory to the Acc register.

Steps 330-335: Extract data 245 from the extracted contents of Acc.

Step 340; R10 the extracted data 245
Set in register.

Now, if we consider a case where the TCB configuration is changed for each system type and stores only the information necessary for that system type, 1 byte of data 245 shown in Figure 2 is extracted. Program 300 changes the capacity of TCB 110,
By changing the position of one word (32 bits) including data 245 and data 245, at least step 315,
It is necessary to change four locations: 325, 330, and 335.

In this way, conventionally, if the configuration of the TCB was different for each system type, it would be necessary to change all the programs that access it. In other words, it is necessary to create a program that accesses the same data for each system type, and to avoid this, the TCB has the same configuration regardless of the system type. Therefore, in some types of systems, there has been a problem in that the capacity of the TCB, that is, the capacity of the operating system occupying the main memory, increases unnecessarily.

An object of the present invention is to solve the above-mentioned problems, and the TCB existing in the main memory has a structure that differs depending on the system type and stores only necessary information. The purpose of this invention is to provide a control method for an electronic computer that can be viewed as a virtual TCB with a configuration and that does not require program changes even if the system configuration changes.

The reason why the program needs to be changed is that the program itself is directly aware of the location address and size of the data it is trying to access. The present invention has focused on this point, and in the conventional TCB configuration, a number is assigned to each piece of data, and a control table is provided that stores the data location address and size corresponding to the number unit, and the program includes TCB for each system type by providing instructions that can access data by specifying only the number.
Even if the configuration is changed, the above object is achieved by simply preparing a control table storing addresses and sizes for each system type, and eliminating the need to change the main body of the program.

Embodiments of the present invention will be described below with reference to FIG. 3 and subsequent figures. The operating system 10 on the main memory 1 has different configurations depending on the system type.
A control table that assigns consistent numbers starting from 1 to the TCB 100 and the data within the TCB, and stores the location address, size, etc. of data corresponding to each number unit:
FDSCR600, TCB100 and FDSCR6
A control table: TDSCR 500 is provided which stores the start address of 00, the capacity of one case of TCB, etc., and the fixed address 400 points to the start address of TDSCR 500. Here, FDSCR600 is
Number assigned to data (called field number) =
1 FDSCR605, field number = 2
FDSCR610, field number = max for
FDSCR615 are arranged sequentially in the order of field numbers, and TDSCR500 is a consistent number (referred to as table number) from number 1 assigned to the management table of the operating system to which the present invention is applied, TDSCR505 for table number = 2 TDSCR510 for, table number =
Arrange sequentially in order of table number, such as TDSCR515 for max. Decide that the TCB table number is number 1.

The processing device 6015 includes a register group 6045,
A calculation mechanism 6030 consisting of a calculation unit 6050, a control mechanism 6035, a control memory 6040, and a memory interface mechanism (memory address register: MAR6020, memory data register:
MDR6025). Processing device 6
The operation of 015 is generally as follows, as in the conventional case.

In response to a command from the control mechanism 6035, the address of the next instruction to be executed is set in the MAR 6020 via the arithmetic mechanism 6030, and by issuing an access request to the main memory 1, the next instruction is transferred from the main memory 1. The instruction to be executed is retrieved and MDR602
Set to 5. The instructions set in MDR 6025 are immediately sent to control mechanism 6035, where they are decoded and control is transferred to a microprogram stored in control memory 6040 that executes the instructions. The micro program is
The control mechanism 6035 sequentially controls the control memory 60.
40 and decoded. Control mechanism 603
5 controls the arithmetic mechanism 6030, memory interface, and memory control mechanism 6010 based on this.

The data manipulation instructions according to the present invention are executed by the microprograms 700 and 800, and are executed by the control memory 60.
Store it in 40.

Figure 4 shows the details of TDSCR. 550 is the start address of the management table to be controlled; 56
0 is the start address of the FDSCR table that stores the data address and size of the management table to be controlled, 565 is the capacity (number of bytes) of one case of the management table to be controlled, and 570 is to be controlled. Indicates the number of cases in the management table.

Figure 5 shows details of FDSCR. 650 is information indicating whether the data is bit data or byte data, 660 is the size of the data (the number of bits if it is bit data, the number of bytes if it is byte data), and 670 is the table of the data. Indicates the relative address (bit address for bit data, byte address for byte data) from the start (start of each case).

Using such a control table, the data to be accessed is managed by number, and on the other hand, instructions as shown in FIGS. 6 to 10 are provided to the program.

FIG. 6 shows an instruction format 700 of GETAR, an instruction for finding the address of a table in which data to be accessed exists. 705 is an operation code (OPCD) indicating that the instruction performs the above function, 710 is the table number (TBLN) of the table in which the data to be accessed exists, and 715 is the case number of the specified table in which the data to be accessed exists. The case number (CSEN) 720 indicating whether the address exists indicates the number (REGN) of the register storing the obtained address.

Figure 7 shows the instructions that actually read data from the table and store it in the table; LDTF/STTF
The instruction format 800 of FIG. 805 is an operation code (OPCD) indicating that the instruction performs the above function; 810 is the table number (TBLN) of the table in which the data exists; 81
5 is the field number (FLDN) of the data to be read or stored, 820 is the number (REGN) of the register storing the address of the table where the data exists (address obtained by the GETAR instruction), and 825 indicates the number of the register that stores the read data when reading data (LDTF), and indicates the number of the register that stores the data (REGN) when storing data (STTF).

8 to 10 show the processing contents of the GETAR, LDTF, and STTF commands.

FIG. 8a shows a processing flowchart of the LDTF command, and FIG. 8b shows the data manipulation contents of the LDTF command. In the following, the processing flow of the LDTF instruction will be explained in accordance with FIG. 8a and FIG. 8b.
This will be explained in relation to the HARD mechanism.

For other commands; STTF and GETAR,
Since this can be inferred from the explanation of this LDTF command, we will simply explain the processing flow.

A(xxx) in FIG. 8b indicates the address of XXX.

Block 1005; The instruction TBLN 810 taken out by the control mechanism 6035 and the memory control mechanism 6
The starting address of the TDSCR 500 obtained through 010 is calculated by the arithmetic unit 6030, and the obtained address 55 of the corresponding TDSCR is sent to the register group 60.
Store in REGi of 45. 51, 52, 53, 54, 55 Block 1010; The control mechanism 6035 controls the contents of REGi of the register group 6045, that is, the corresponding
The address of TDSCR is accessed via the memory control mechanism 6010, and TPFDS of TDSCR500 is accessed.
560 to obtain the start address of the FDSCR 600.

Block 1015; The control mechanism 6035 outputs the fetched instruction field FLDN815 and the start address of the previously obtained FDSCR600 to the arithmetic mechanism 6.
030, and the address of the corresponding FDSCR
Get 59. The obtained address 59 is assigned to register group 6.
Store in REGj of 045. 56, 57, 58, 59 block 1020;, the control mechanism 6035 controls the contents of REGj of the register group 6045, that is, the corresponding
Address 59 of FDSCR to memory control mechanism 6010
Access via and in FDSCR600
DATATYPE 650 determines whether the field data is byte data or bit data.

Block 1025: If the field data B150 is byte data, the main memory address of the corresponding table body 100 indicated by the instruction field REGN820 retrieved by the control mechanism 6035 and the DATAADDR in the corresponding FDSCR indicated by REGj.
670 is added by the arithmetic unit 6030, and the main memory address 61 of the corresponding field data B150 is
get. The obtained address 61 is the register group 60
Store it in REGk in 45. 60, 61 Block 1030; From the main memory address of the field data B150 indicated by REGk,
The data for DATASIZE 660 bytes in the FDSCR 600 is accessed via the memory control mechanism 6010 to obtain the corresponding field data B150. 62, 63 Block 1035; The obtained data B63 is input to the instruction field taken out by the control mechanism
Store it in REGD825 and end the process. (REGD
(1 register of register group 6045) 64, 65 block 1040; If the field data A 160 is bit data, first register group 604
Of address 59 of the FDSCR indicated by REGi of 5,
The contents of the DATAADDR section 670 are obtained via the memory access mechanism 6010 and multiplied by 1/8 by the arithmetic mechanism 6030 to obtain the byte address of the field data A160.
Instruction field REGN8 retrieved by 35
The main memory address 20 of the corresponding table body 100 is further added by the calculation mechanism 6030 to obtain the main memory address 61 of the corresponding field data A160. The obtained address 61 is stored in REGk in the register group 6045. 60, 61 Block 1045; Main memory address 61 and subsequent 4 of the field data A160 indicated by REGk
Bytes of data are accessed via the memory control mechanism 6010, and the corresponding field data A16 is
Obtain 4 bytes of data 67 including 0. 66, 67 Block 1050; Field data A160 obtained by obtaining the contents of the DATAADDR section 670 from the address 59 of the FDSCR indicated by REGj via the memory access mechanism 6010 and multiplying it by 1/8 by the arithmetic mechanism 6030. bit address (1/
The remainder after multiplying by 8 becomes the bit address),
DATASIZE660 (bit size in case of bit data) in FDSCR determines data A among the 4-byte data 67 including the data A obtained earlier.
Only 160 is cut out by the calculation mechanism 6030.
68 Block 1055; The obtained data A160 is shifted to the right by the arithmetic mechanism 6030, and the instruction field taken out by the control mechanism 6035 is
Store it in REGD825 and end the process.
(REGD is 1 register of register group 6045)
69, 70 FIG. 9 is a processing flow diagram of the GETAR command.
First, the start address of the TDSCR corresponding to the specified TBLN is determined (step 905). I asked for the next
Calculate the relative address from the beginning of the table to the specified case from SIZE in TDSCR and the specified CSEN (calculated by SIZEX (CASE - 1))
(Step 910). Next, the relative address obtained above and the table top address in the TDSCR are added to obtain the absolute address for the specified case (step 915). As described above, the absolute address of the designated case in the designated table is determined, and the process is completed by storing it in the designated register (step 920).

FIG. 10 is a processing flowchart of the STTF command.
Steps 1105 to 1120 determine whether the data is bit data or byte data using the same procedure as the LDTF command. If it is byte data, add the address of the specified table (contents of REGN) and DATTADDR in FDSCR to find the byte address of the data 1125. The information specified by REGD is stored for the number of bytes specified by DATASIZE in FDSCR after the address of the data obtained above (1130), and the process ends. In the case of bit data, find the byte address part from DATAADDR (bit display) in FDSCR,
The address indicated by REGN is added to obtain the byte address of the data 1135. After the byte address obtained above, 4 bytes of information are temporarily read out to the work area within the instruction. in FDSCR
The bit position of the data is determined using DATAADDR and DATASIZE, and the data portion is erased from the read information (1145). Embed the write data specified in REGD in the area erased above1
150. The obtained 4-byte data is stored again starting from the byte address at which it was read out (1155), and the process ends.

By assigning numbers to the data to be accessed as described above, providing a control table that stores the address and size of the data, and providing an instruction to access the data by specifying the number, it is possible to achieve the same result as shown in FIG. A conventional program 300 for reading one byte of data is a program 60 shown in FIG.
It becomes 00. Here, even if the position and size of the 1-byte data in Figure 1 are changed, the DATASIZE and
By simply changing DATAADDR, the same data can be accessed without changing the program 6000 shown in FIG. 11.

Now, create the basic mechanism as above,
The TCB has the following configuration.

FIG. 12 shows the configuration of the TCB 1500 as seen by the accessing program. In the conventional TCB configuration 110 shown in FIG.
Assign a field number from 1 to 44 to each piece of data. Here, as explained in FIG.
(Field numbers = 24 to 30, 31, 33, 34) and 270 (Field number = 40) are necessary information only for the system with DISC type, and 210
(Field number = 18) and 220 (Field number = 22) and 250 (Field number =
44) is information required only in a multi-computer system format. The TCB visible to programs accessing in this way is the same regardless of the system type. This TCB1500
does not exist in main memory, but in the virtual TCB (VTCB)
It is called.

Existing in main memory in Figures 13 to 15
The configuration of the TCB (this TCB is called RTCB) is shown below.

Figure 13 shows RTCB in system configuration with DISC.
Show the configuration. From the VTCB in FIG. 12, information necessary only for the multi-computer system configuration (field numbers = 18, 22, 44) is removed, resulting in a configuration of 32 bytes per case.

Figure 14 shows the RTCB configuration in multi-computer system mode, and this RTCB is VTCB
Same configuration as .

Figure 15 shows the main memory only system configuration.
Shows RTCB configuration 3000, from VTCB
Information required only for system configurations with DISC and multi-computer system configurations (field numbers = 8, 12 to 17, 20, 24 to 30, 31, 33, 34, 40,
18, 22, 44) are removed to make one case 26 bytes.

FDSCR is created for each RTCB. 1st
Figure 6 shows the FDSCR in the system configuration with DISC. FDSCR3 for field number (FLDN) = 1
005, = FDSCR3010 for =2, etc.
Create up to FDSCR3050 for 44 in order, and each
The FDSCR contents are created to represent RTCB・TYPE1 (Figure 13). For example, FLDN=42
The FDSCR 3040 is created such that the data type (DATATYPE) is byte data, the size (DATASIZE) is 1 byte, and the address (DATAADDR) is the 29th byte. There is this RTCB on VTCB.
Data not found in TYPE1 (FLDN=18, 22, 44)
FLDN3020, 3030, 3050 for
The content is set to "0" to clearly indicate that there is no data for the FLDN in this RTCB.

FIG. 17 shows the FDSCR 3100 in the form of a multi-computer system. From 3105 for FLDN=1 as well as RTCB・TYPE1 (Fig. 16)
They are created in order up to 3120 for FLDN=44, and the contents of each are created to represent RTCB・TYPE2 (FIG. 14).

Figure 18 shows the main memory only system configuration.
Shows FDSCR3200. Similar to the former, FLDN=
Created from 3205 for 1 to 3325 for FLDN=44, and the contents of each are RTCB・TYPE3 (Figure 15)
Create it to represent. For example, for FLDN=42
FDSCR3320 has DATATYPE and
DATASIZE is the same as the former, but
DATAADDR is the 23rd byte. Data that exists on VTCB but does not exist in this RTCB・TYPE3 (FLDN=8, 12-18, 20, 22, 24-31, 33, 34,
40, 44) FDSCR3215, 3220~
3250, 3255, 3260, 3265-33
00,3305,3310,3315,3325
The content is set to "0" to clearly indicate that there is no data for the FLDN in this RTCB.

Figure 19 shows the FDSCR for each system type.
Shows the relationship of RTCB. The TCB start address (TPTCB) 4000 in TDSCR505 for TCB is
Depending on the system type, it refers to RTCB/TYPE1 or 4005, or it refers to RTCB/TYPE2 4010,
Or refer to RTCB・TYPE3 or select 4015 at the time of system generation. Similarly
For the TPFDS 5000 in the TDSCR 505, either 5005, 5010, or 5015 is selected during system generation depending on the system configuration.
SIZE565 in TDSCR505 is RTCB/
If TYPE1 is selected, a value of 32 is embedded, if TYPE2 is selected, a value of 37 is embedded, and if TYPE3 is selected, a value of 26 is embedded during system generation.

As described above, by creating an FDSCR for each system type, selecting the FDSCR and RTCB that match the system type during system generation, and accessing programs with only the VTCB configuration in mind, you can create an FDSCR for each system type without changing the program.
It becomes possible to change the configuration of TCB.

In this embodiment, in the case of a main memory only system with the smallest RTCB, the TCB capacity can be reduced by 11 bytes per task compared to the conventional TCB capacity.

The effects of the present invention are shown below.

1. It is now possible to change the TCB configuration within the operating system that manages tasks without affecting the access program.

2. Along with this, it is possible to suppress the increase in the TCB capacity, which has increased unnecessarily depending on the system configuration, and furthermore, the increase in the capacity of the operating system that occupies main memory, expanding the range of main memory that can be used by users. became possible. (Depending on the system configuration, up to 11 bytes can be reduced per task, and with the conventional maximum number of tasks of 255, the capacity can be reduced by 2805 bytes.)

[Brief explanation of the drawing]

Figure 1 is a conventional TCB configuration diagram, Figure 2 is a conventional TCB configuration diagram.
Example of a TCB data access program, FIG. 3 is an overall view of a table created in relation to the present invention, FIGS. 4 and 5 are diagrams showing details of FIG. 3, and FIGS. 6 and 7 are created in relation to the present invention. 8 to 10 are instruction processing flowcharts, FIG. 11 is an example of a TCB data access program according to the present invention, and FIGS. 12 to 15 are configuration diagrams of a TCB modified according to the present invention. Figures 16 to 19 are
FIG. 3 is an example diagram of a table provided in connection with the present invention. 500...TDSCR, 600...FDSCR, 700
...GETAR instruction, 800...LDTF/STTF instruction,
1500...Virtual TCB, 2000...RTCB・
TYPE1, 2100...RTCB・TYPE2, 220
0...RTCB・TYPE3, 3000...FDSCR・
TYPE1, 3100...FDSCR・TYPE2, 320
0…FDSCR・TYPE3.

Claims (1)

[Scope of Claims] 1. A method for controlling an electronic computer in which a plurality of tasks, which are units of work generated from a user program, are stored in a main memory and are operated on an arithmetic processing unit, comprising the steps of: A task management table that stores a plurality of data necessary for controlling the data and has one type of data arrangement configuration among a plurality of types of data arrangement configurations according to the types of the plurality of data to be stored,
A plurality of data are stored for each of the plurality of tasks on the operating system of the main memory, and the data related to the number is stored in correspondence with a number commonly assigned to each of the data in the plurality of types of task management tables. a control table storing arrangement information in each of the task management tables for each of the plurality of types of data arrangement configurations is stored on the operating system; A method for controlling an electronic computer, comprising: executing an instruction for obtaining the arrangement information by referring to the control table for a data arrangement configuration, and an instruction for specifying the arrangement information and accessing the data. 2. The plurality of types of data arrangement configurations correspond to the types of the plurality of data types in each of the plurality of system configurations when using the arithmetic processing unit, the main memory, or the auxiliary storage device. Method for controlling an electronic computer as described in Section 1. 3. The plurality of system types include a first system type that uses all of the arithmetic processing unit, the main memory, and the auxiliary storage device, a second system type that uses only the arithmetic processing unit and the main memory, The method of controlling an electronic computer according to claim 2, which is a third system type using a plurality of computers.