JPH10187586A - Peripheral device and information processor including the same peripheral device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily connect many peripheral devices by setting an address range which is accessed with the same ID for each peripheral device and making each peripheral device respond only within the corresponding address range. SOLUTION: Disk drives 2 as a master peripheral device and one or plural slave peripheral devices having the same ID are connected through a bus; and the disk drive 2 as the master peripheral device holds an upper-limit address responding to access and the disk drives 2 as the slave peripheral devices hold specific address ranges which exceed the said upper-limit address and do not overlap with each other. Thus, the address ranges accessed with the same ID are set for the respective disk drives 2, which respond only within the corresponding address ranges to eliminate the limit of the number of connections with IDs given to the disk drives 2, so that many disk drives 2 are easily connected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a peripheral device for accessing a plurality of peripheral devices by connecting them to a bus, and an information processing apparatus including the peripheral devices.

In recent years, peripheral devices can be connected to a bus by a predetermined number of different IDs. For example, in SCSI, only seven IDs (one host) can be connected, but the number of IDs is further limited. It is desired to connect many peripheral devices without the need.

[0003]

2. Description of the Related Art A conventional bus, for example, SCSI, is shown in FIG.
As shown in (a-1), there is an ID for a device connected to the SCSI, and the specification is such that a bus width of 8 bits and eight devices can be connected. Therefore, excluding the host, the device is 7
For example, in the case of a disk device of 100 Mbytes, only a total of 700 Mbytes could be connected.

Further, as shown in FIG. 10B, even when a disk device is connected in parallel from a host, if the number of IDs is limited to seven except for the host, the disk device is not connected. Only a total of seven could be connected.

[0005]

The above-mentioned conventional FIG.
In the SCSI of (a-1) or the bus of FIG. 10B, an ID is assigned to a device to be connected, and only seven disk devices can be connected except for a host. there were.

Further, as shown in FIG.
Sub IDs are provided for one ID (in the case of SCSI, LUN
(Called Logical Unit Number)) There is a method of increasing the number of connectable single IDs such as sub ID “LUN = 0” and “LUN = 1”. However, there is a problem that extra cables and connectors are required in addition to the connection to the disk controller. Also, LUN is
There is a problem that only eight units can be connected even though they are represented by three bits.

[0007] The present invention solves these problems,
Address range accessed within the same ID (upper / lower)
Is set for each peripheral device, each peripheral device accesses only when the address is within a corresponding address range, and the object is to realize a large number of connections without limiting the number of connected devices by the ID assigned to the peripheral device.

[0008]

Means for solving the problem will be described with reference to FIG. In FIG. 1, a host (upper device) 1 accesses a peripheral device such as a plurality of disk devices 2.

The disk device 2 is one of the peripheral devices and is accessed from the host 1.

Next, the operation will be described. The disk device 2, which is a peripheral device, performs processing in accordance with data from the bus only when the ID received from the bus is the same as the ID held by itself and falls within the address range held by the received address. I have to.

At this time, the peripheral device of the master having the same ID and the disk device 2 as the peripheral device of one or more slaves access the disk device 2 as the peripheral device of the master via the bus. The upper limit address for responding is held, and the disk device 2, which is the peripheral device of each slave, holds a predetermined non-overlapping address range exceeding the upper limit address for responding to access.

Therefore, an address range (upper limit / lower limit) to be accessed within the same ID is set for each disk device 2 as a peripheral device, and each disk device 2 responds only when it is within the corresponding address range. In addition, it is possible to eliminate the limitation on the number of connections by the ID assigned to the disk device 2 as a peripheral device, and to easily connect a large number of disk devices.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments and operations of the present invention will be sequentially described in detail with reference to FIGS.

FIG. 1 shows a system configuration diagram of the present invention.
In FIG. 1, a host 1 accesses a plurality of serially connected disk devices 2, and here has an ID = 7.

The disk device 2 is connected to the host 1 in a serial manner, and comprises ID information 3, address information 4, control means 5, a disk 6, and the like.

The peripheral device includes a disk device such as a hard disk device and a magneto-optical disk device, and an input / output device such as a CD-ROM drive device and an image scanner device.
In the embodiment, a disk device will be described below as an example.

The ID information 3 is information such as an ID assigned to the host 1 and each disk device 2. In this case, the ID shown in FIG.
Are given.

The address information 4 is information relating to an address range to which a response is made when accessing from the host 1, and includes an upper limit address for a master and upper and lower limit addresses for a slave.

The control means 5 performs various kinds of control, and controls the host 1 based on the ID information 3 and the address information 4.
This is to determine whether or not to respond to the access from the user, or to access the disk 6 when it is determined to respond.

The disk 6 is a medium for storing data, for example, a hard disk. FIG. 2 is a configuration diagram of a main part of the disk drive of the present invention.

In FIG. 2, the disk device 2 has an SCS
I is connected to a daisy chain and writes data to the disk 28 in accordance with the access from the host 1,
Read out and send it to the host, etc.
21 to 28 and the like.

The connector 21 is a connector for connecting to the SCSI, for daisy-chain connection, and for connecting a terminating resistor when it is terminated.

The SCSI controller 22 controls transmission and reception of data to and from SCSI. The ID setting unit 23 sets ID and master / slave information, and is, for example, a short plug shown in FIG. 3 described later or a non-volatile memory (not shown). In the case of a short plug, the setting is manually performed by an operator.
On the other hand, in the case of the non-volatile memory, the ID and master / slave information are set in a state where the command is sent from the host 1 to the disk device 2 and the setting mode is set. In this state, the ID and master / slave information are set.

The disk controller 24 controls the disk 2
8 for writing data and reading data from the disk 28. The CPU 25
Disk device 2 according to the program stored in ROM 27
Of various controls.

The RAM 26 is a memory for temporarily storing data and the like. The ROM 27 is a nonvolatile read-only memory for storing programs and data.

The disk 28 is a hard disk for storing data. Next, setting of ID information will be described with reference to FIG. FIG. 3 shows the ID of the disk device of the present invention.
The example of information (the 1) is shown. This is the ID of FIG.
= 0 and corresponds to the ID setting unit 23 of FIG. 2, and explains the setting of ID and master / slave distinction.

FIG. 3A shows the state of a shorting plug on the printed circuit board of the first master disk device 2. The master disk device 2 is set as shown below.

ID jumper plug: ID = 0 Master / slave jumper plug: not shorted (representing the master) FIG. 3B shows a jumper plug on the printed circuit board of the first slave disk device 2 The state of is shown. The first slave disk device 2 is set as shown below.

ID jumper plug: ID = 0 Master / slave jumper plug: 0 (represents the first slave) FIG. 3C shows the printed circuit board of the second disk device 2 of the slave. The state of the short plug is shown. The second slave disk device 2 is set as shown below.

ID jumper plug: ID = 0 Master / slave jumper plug: 1 (representing the second slave) By setting the above jumper plugs, the master disk device 2 ( (FIG. 3 (a)) First slave disk device 2 (FIG. 3 (b)) Second slave disk device 2 (FIG. 3 (c)) That is, it has been set to that effect. Then, as described in FIG. 5, address ranges different from a total of three for the master disk device 2, the first slave disk device 2, and the second slave disk device 2 are used as address information 4. Give.

FIG. 4 shows an example (No. 2) of the ID setting section of the disk device of the present invention. This is an example of a circuit in which the CPU 25 reads the ID information of the short flag and the master / slave information set as shown in (a), (b), and (c) of FIG. 5 is a circuit example for reading the contents of a unit 23.

FIG. 4A shows an example of a decoder circuit.
You. CPU 25 sets address ADRS 01 to ADRS
15 is input to the decoder circuit shown in FIG.
When EAD is input, ID read signal * READ I
D or master / slave read signal * READ
One of the MS signals is decoded and transmitted.

FIG. 4B shows an example of a SCSI ID reading circuit. This is because the ID read signal * READ decoded in FIG. ID is LS2 of IC element
When the signal is input to the enable terminal EN of the ID 44, whether the ID short plug 0, 1, 2 is short-circuited and grounded is determined via the LS 244 via the internal processor data bus DATA. 00 or DATA 02
, And the CPU 25 reads and recognizes the ID.

FIG. 4C shows an example of a master / slave reading circuit. This is because the master / slave read signal * READ decoded in FIG. When MS is input to the enable terminal EN of the LS 244 of the IC element, it is determined which of the master / slave jumper plugs 0, 1, and 2 is short-circuited and grounded.
DAT of internal processor data bus via LS244
A 00 or DATA 02 respectively, and the CP
U25 reads and recognizes master / slave distinction.

As described above, (a), (b),
As shown in (c), it is possible to read and recognize the state of the short plug (ID and master / slave distinction) set in the ID setting unit 23 in each disk device.

FIG. 5 shows an example of address information according to the present invention.
This is an example of the address information when the ID information 3 of FIG. 3 is set, and explains an example of the setting of the ID of FIGS. 1 and 2.

FIG. 5A shows a general expression of a setting command. Here, the settings are made as shown below. 1st byte: A code representing a setting command is set.

Second byte: Number of connected devices (sets the number of disk devices connected to the same ID) Third byte: First byte of the address range of the first device (master) Fourth byte: First device 2nd byte of (master) address range 5th byte: 1 of 2nd (slave) address range
Byte • Byte 6: Address range 2 of the second device (slave)
Byte 7th Byte: 1 of address range of 3rd (slave)
Byte 8th Byte: 2 of address range of 3rd (slave)
Byte No. FIG. 5B shows a specific expression of the setting command.
Here, in contrast to the general expression of FIG.
Are set specifically as shown below in accordance with the ID information. Here, the first (master) disk device 2 is 500 MB and the second (first slave) disk device 2 is 800 MB.
The disk unit 2 of the third MB (second slave) is 400
MB.

1st byte: 'FF' (currently undefined F
F is set as a setting command) 2nd byte: '03' (represents 3 connected units) 3rd byte: '05' 4th byte: '00' (3rd and 4th bytes are '050')
0 "(represents 500 MB) 5th byte: '08' 6th byte: '00'('080 at 5th and 6th bytes)
0 "(represents 800 MB) 7th byte: '04' 8th byte: '00'('040 at the 7th and 8th bytes)
0 "(400 MB)) By setting the ID information 3 and the address information 4 in FIG. 3 and FIG. 5, three disk devices can be set in different address ranges for ID = 0. Only in the case of the above, each disk device 2 responds in response to the access from the host 1. The operation will be sequentially described in detail below.

FIG. 6 is a diagram for explaining the operation of this embodiment. Here, the host 1 corresponds to the host 1 in FIG. 1, and the master disk device 2 and the slave disk device 2 are the master disk device 2 with ID = 0 and the slave disk device 2 with ID = 0 in FIG. Corresponding.

In FIG. 6, the host 1 arbitrates for S1. This is because a host 1 in FIG. 1 serially connects a plurality of disk devices 2 (here, S
Acquire exclusion of CSI) (After confirming that the busy signal is inactive, activate it to acquire exclusion).

In step S2, selection ID = 0 is performed. This means that ID = 0 is sent to the bus, and the master of the corresponding disk devices 2 returns a response in S2, which is received and confirmed by the host 1. At this time, the slave sets the ID =
It recognizes that 0 itself has been selected (only the master returns a response, not the slave).

By the above S1 to S3, it can be recognized that the master and the slave with the ID = 0 are selected by the selection of the host 1 with the ID = 0.
In S4, the host sends a read command.

In S5, the master determines the address length in response to the transmission of the read command in S4. On the other hand, S6 corresponds to the transmission of the read command of S4,
The slave determines the address length.

In S7, the first half master recognizes the response.
On the other hand, in S8, the second half slave recognizes the response. S9
In this case, the master performs head selection and seek operations. As a result, the master has read the first half of the data from the disk 6. On the other hand, in S10, the slave performs head selection and seek operations. Thus, the slave has read the latter half of the data from the disk 6.

In step S11, a reselection phase for performing bus reconnection is performed. S12 returns a response to the bus reconnection of S11. In S13, the master transfers data (first half data) to the host.

In S14, the master releases the bus. S1
In 5, the slave transfers data (second half data) to the host. S16 receives the data transferred in S13 and S15 and returns a response.

In S17, since the slave has finished transferring all data, the status is transmitted to the host. S1
In step 8, the host receives the status of S17 and returns a response.

By the above S4 to S18, two of master and slave having ID = 0 selected from the host
In response to the read command, the master reads data from the disk 6 and transfers the data to the host in the first half, and the slave transmits data to the disk 6 in the second half.
And a data transfer to the host, a series of ID = 0
Has been read from the disk device.
This allows the host 1 to be conscious of the number of disk devices with ID = 0 even if the number of disk devices 2 with ID = 0 is one or two (master and slave) and three or more. This makes it possible to access data without any restrictions and to expand the accessible capacity.

Next, the operation for reading data from the disk device 2 with ID = 0 (master) and the disk device 2 with ID = 0 (slave) connected to the SCSI will be described with reference to FIGS. This will be described in detail. This is when data is read from the disk device 2 with ID = 0 (master) and the disk device 2 with ID = 0 (slave) in FIGS. 1 and 2.

FIG. 7 is a time chart (part 1) of the present invention.
Is shown. This is a time chart when a master and a slave with ID = 0 are selected from the host and a READ command, address data, and length data are stored (this is a time chart of processing performed as preprocessing in FIG. 8). .

FIG. 7 shows an arbitration phase. This is a phase in which the host 1 confirms that the * BUSY signal is inactive (H level) and then activates (L level) to acquire SCSI.

Is a selection phase. This means that after Host 1 has acquired SCSI at
This is a phase in which ID = 0 is transmitted and a master and a slave having ID = 0 are selected.

Is a READ command storage phase. This is because the host 1 sends the READ command to the SCSI
, And the master and slave selected in step 1 respectively store the READ command.

Saves address data. That is, following the READ command, the host sends address data (for example, a head address read from the disk 6) to the SCSI, and the master and the slave store this.

Saves length data. This means that, following the address data, the host sends length data (for example, the length of reading the disk 6 from the head address) to the SCSI, which is stored by the master and the slave.

As described above, after the host acquires the SCSI and sends out the ID and selects the master and slave shown in the figure, the READ command, the address data and the length data are sequentially sent out to the SCSI and the master and slave are sent to the master and slave. It can be saved. And
A series of reading can be performed by referring to FIG. 8 to be described later.

FIG. 8 shows a time chart (No. 2) of the present invention. This is a time chart when data is read from the master and the slave with ID = 0 from the host, and is a process performed after FIG.

FIG. 8 shows a state where the host 1 has read data from the master. This means that the host 1 acquires the SCSI arbitration (* BU shown in the figure).
SY is activated when it is low), the master and the slave shown in the figure with ID = 0 are selected, the read command is transmitted, and the host takes in the first data from the master (S1 to S7 in FIG. 6 described above). Equivalent).

The host fetches the following data in the same manner as in (1). Is a state in which the host has taken in data in the address range (here, the upper limit of the address) set as the master address information.

"1" indicates a state in which the host 1 has read data from the slave. In this method, the slave sends data of the address upper limit set as the address information to the bus, and the data is taken in by the host. At this time, the master releases the bus so that the slave does not collide with the transmission of data (corresponding to S14 in FIG. 6 described above).

In the same way, the host takes in the following data. , The host captures the exit status. This means that the slave sets data in the address range (here, the upper limit of the address) set as address information to the SCS.
Since transmission to I has been completed, an end status indicating that a series of data transfer has been completed is transmitted to SCSI, and the host takes in the data to recognize that a series of read processing has been completed.

As described above, the master and the slave having the ID = 0 send data to the SCSI up to the upper limit address set as the address information of the master in the phase (3). After transmitting data from the lower limit address to the upper limit address set as information to the SCSI, an end status is transmitted to the SCSI, and the host takes in these, thereby completing a series of read processing. At this time, the host 1 fetches data from the SCSI without being aware of the existence of the master and the slave, receives an end status at the time of termination, and can recognize that a series of read processing has been completed.

FIG. 9 shows a serial / parallel connection configuration diagram of the present invention. FIG. 9A shows a serial connection form diagram. This is a connection configuration diagram when seven disk devices are serially connected to the SCSI described in FIGS. 1 and 2, and a terminating resistor is connected to the end. In the case of this SCSI, the ID is 3 bits and the number of IDs is a total of 8 from 0 to 7, but since ID = 7 is assigned to the host,
Seven IDs from ID = 0 to ID = 6 are allocated to the disk devices 2, respectively. Here, as shown in FIG.
And master / slave, and set ID = 0 as master,
A total of three disk devices, slave 1 and slave 2, are allocated.

FIG. 9B shows a parallel connection configuration diagram. This is different from the case where seven disk devices are serially connected to the SCSI described with reference to FIGS. 1 and 2, but the case where the disk devices 2 are connected from the host to parallel buses, respectively. In the case of this parallel bus, as in the case of the SCSI in FIG. 9A, the ID is 3 bits and the number of IDs is 0 to 7, that is, a total of 8 IDs.
D = 7 is assigned, and seven IDs from ID = 0 to ID = 6 are assigned to the disk devices 2, respectively. Here, ID and master / slave are set as shown in FIG.
A total of three disk devices of master, slave 1, and slave 2 are assigned to ID = 0, and each is connected to a parallel bus.

FIG. 9 (c) shows a configuration in which one bus is connected in parallel. This is because, in addition to the serial / parallel connection configuration shown in FIGS. 9A and 9B, the disk devices 2 are connected in parallel to one bus, respectively, and are serially connected to the SCSI described in FIGS. As in the case of connecting two disk devices, the ID is 3 bits and the number of IDs is 8 in total from 0 to 7, and ID = 7 is assigned to the host,
Seven IDs from ID = 0 to ID = 6 are allocated to the disk devices 2, respectively. Then, as shown in FIG. 3, an ID and a master / slave are set, and a total of three disk devices of master, slave 1, and slave 2 are assigned to ID = 0 and connected to one bus.

[0067]

As described above, according to the present invention,
Address range accessed within the same ID (upper / lower)
Is set to each peripheral device, and each peripheral device responds only when it is within the corresponding address range. Therefore, the limitation on the number of connections by the ID given to the peripheral device 2 is eliminated, Can be easily connected. Thus, the capacity can be increased many times by connecting a plurality of peripheral devices with the same ID.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of the present invention.

FIG. 2 is a configuration diagram of a main part of the disk device of the present invention.

FIG. 3 is an example (No. 1) of an ID setting unit of the disk device of the present invention.

FIG. 4 is an example (No. 2) of an ID setting unit of the disk device of the present invention.

FIG. 5 is an example of address information according to the present invention.

FIG. 6 is a diagram illustrating the operation of the present invention.

FIG. 7 is a time chart (No. 1) of the present invention.

FIG. 8 is a time chart (2) of the present invention.

FIG. 9 is a serial / parallel connection configuration diagram of the present invention.

FIG. 10 is an explanatory diagram of a conventional technique.

[Explanation of symbols]

 1: Host 2: Disk device 3: ID information 4: Address information 5: Control means 6: Disk

Claims (2)

[Claims] 1. An apparatus for transmitting or receiving data to a higher-level device via a bus, wherein an ID received from the bus is the same as an ID held by itself and falls within an address range held by a received address. A peripheral device comprising means for performing processing in accordance with data from a bus only when there is a need. 2. The master peripheral device having the same ID and one or a plurality of slave peripheral devices are connected via a bus, and an upper limit address for responding to access to the master peripheral device is set. 2. An information processing apparatus including a disk device according to claim 1, wherein a predetermined address range that does not exceed the upper limit address in response to access to the peripheral device of each slave is accessed.