JPS61125254A - Ring-status interrupting system - Google Patents

Abstract

PURPOSE:To improve the overall putthrough and to relieve the load of the CPU by so forming the system that each node transmits a frame to itself in order to test its state of coupling with the ring and by checking its result by a dedicated LSI, and further generating an interruption in the firmware based on the above result. CONSTITUTION:The transmission frame 500 from the node A goes around a ring, and is inputted to the receiving circuit of a TPC (token passing controller) 10. If the inputted byte is a source address part SA, the TPC10 compares it with its own address (MA). And the comparison-result is expressed in the third bit and the second bit of a receiving status register RXFS1028. If it is 00, then SA=MA. At this time, the outputs of an inverting circuits 1034 and 1035 shown in figure (b) become a logic 1. In case when the ring turn-around test is made valid, an FF1036 is set by an MUP20. An FF1037 is a storage element to show whether or not the status is valid, and shows that the status information is exactly set in an RXFS. Accordingly, if the circuit shown in figure (b) is used, when the bits 1036 and 1037 both are logic 1 and hence are in a valid status and if SA=MA, a turn-around interruption FF1039 is set.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an access method to a transmission medium of a local area network (LAN) that connects a plurality of 61 computers by communication in a geographically close area. .

In particular, the present invention relates to a token-based medium access control state transition interrupt system in which only a computer that captures a free token has transmission rights in a ring-shaped transmission medium.

In recent years, the era has shifted from centralized processing by a central processing unit (CPU) to separated processing by multiple workstations. Local area networks (LANs), which organically link computers located in geographically close campuses or buildings through communication networks to efficiently execute two-way distributed processing and office automation, have been attracting attention. By making effective use of this LAN, you can efficiently transfer programs, text, or image data between workstations and three computers, or input/output large-capacity data (input/output to devices with files).■ , Typical AN coupling methods include the lotus type and the ring type.In either the bus type or the ring type, in a LAN (one packet per LAN), packets mainly consisting of data and sending/receiving signals, etc. However, in this case, multiple workstations, called No. 1-, share the transmission medium.

Access to the medium is controlled based on communication rules called protocols.

- It is important to make seven access selections. Therefore.

Basic I11 quality of the network depending on the access method.

In other words, characteristics such as transmission capacity (how many packets can be transmitted per unit time), failure countermeasures, and ease of system design are determined.

[Conventional technology]

Conventionally, this kind of 1. The access method for 8N is C.
3M8/HeC, (Carrier 5ense mul
tiple access/collision det
), time slot method, and token passing method. The C3MA/CF) method is a method typified by Ethernet, in which No. 1 that has a packet to send will immediately send it if the bus is free, and if the bus is in use, it will stop sending and send it as needed until resending. This method waits for a certain amount of time and then sends the message again. One method for time slots is to divide a fixed period of time according to a number of nodes, and each node is allowed to transmit packets only during a determined time slot period, or to allocate time slots according to requests. Finally, in the 1-kun passing method, packets called fleet tokens, which represent transmission rights, are mainly circulated through a ring-shaped transmission medium, and the nodes that want to send wait until a token comes around. −
In this method, a data packet is transmitted by taking in the l·-kun into the transmission medium, and then the l·-kun is sent out again to the transmission medium.

In general, there are two types of coupling methods to the No. 1 transmission medium of LAN: bus type and ring type.
The CD method is mainly used, but the disadvantages of the bus method are that it is not suitable for long-distance or high-speed networks in principle, and that optical fiber is difficult to use.
In particular, as the load increases, delays due to collisions increase and transmission efficiency decreases. Therefore, when the load is large, the l·-Kunling method is said to be the most efficient. That is, it has the advantage that even when the load is increased, the change in delay time does not increase rapidly, the transmission capacity is high, and the maximum delay time is determined by the propagation delay between the rings. However, the disadvantages are that a failure of one disconnection in the 5-ring will affect the whole, packets will continue to circulate, and tokens will be lost, so a function to deal with this is required.

Traditionally, most of this type of token ring media access control has been performed using firmware within each node. Therefore, while the firmware is executing the control to capture the token, transmit the frame data, and send the token again, the firmware cannot perform any other work, and therefore the throughput of each node decreases. Not only that, but it also had the disadvantage of reducing the communication efficiency of the entire system.

[Problem that the invention seeks to solve]

In order to eliminate the above-mentioned drawbacks of the conventional technology, the present invention has the following features:
In the token ring system, state control such as transmitting frame data after capturing l-kun and sending the token again can be performed by dedicated hardware independently of firmware.

This solves the problem of how to separate dedicated hardware and firmware.

[Means to solve the problem]

The present invention is a local area network in which access rights to a ring-shaped transmission medium are controlled using tokens, in which each node transmits a frame addressed to its own node in order to check the state of connection with the ring, and prepares the results. At the point when it was completed.

In other words, when it is possible to check that the source address is the own node, and when it is possible to check whether the FCS error was detected on the own node or on another node, an interrupt is used to notify the internal Bromo Nosa, An object of the present invention is to provide a ring state interrupt method characterized by displaying the contents in a register.

[For production]

The present invention sends a frame to its own node in order to test whether it is connected to the ring, and determines whether the source address in the frame is the own node's address using a dedicated node.
S■ checks the information and generates an interrupt to the firmware based on the result.

(Example〕 Next, embodiments of the present invention will be described with reference to the drawings.

Figure 1 fat, (bl is a block diagram explaining the ring state interrupt method of the present invention, in a local network (1, AN) in which a plurality of computers called nodes are connected to a ring-shaped communication path. , based on the principle of the token method that only the node that captures a free token (hereinafter simply referred to as a "token") has the right to transmit, an automatic Send a frame to node l', and
The dedicated LSI checks whether the address is the own node address,
A method is shown in which an interrupt is generated in the microprogram (firmware) based on the result.

In the case of the token passing method, when a 3-byte L-kun is passed around the ring and a certain No1' captures the L-kun, it has the right to transmit and sends a data frame. There are several methods for determining when to release the 1-kun when the node finishes transmitting, but in this embodiment, the 1-kun is released when the data frame has finished being transmitted. The method is to transfer the transmission right to another No. 1 by transmitting the ``No. 1''. That is, the algorithm releases 1.-kun as soon as it finishes sending out the packet. Regarding the above l-kun passing method, the third
The diagram will be explained. For example, in Fig. 3 fal, if four nodes A, B, C, and D are connected in a ring, free 1--kun is between C and D, and all nodes 1'', A, B This is a repeat state in which the frame can be received and the same data can be streamed even though none of the nodes C, D, and C have captured the i-kun.
Now, when node D has captured the fleet token, node D enters a token holding state, that is, a waiting state for the right to transmit a frame.

At this time, other No.1゛A, l'3. C and D remain in the repeat state. In the same figure fcl, 9 no 1”D
is a state in which a busy token is sent to the ring while continuing to hold a token. Therefore, no FD 1.
- Busy tokens can be sent in the fleet hold state, but other nodes A, B, and C in the repeat state will repeat fleet tokens and busy tokens.

In fd) in the same figure, the state where no~l''r) is about to finish transmitting the frame is in the token sending state. In the same figure (el).

Node D continues to be in the token sending state.

A Node B in a repeat state is receiving frames. In ffl of the same figure, when node D recognizes the frame header while continuing the token sending state,
This is a state in which free tokens are sent out and busy tokens are removed. Then, as shown in the same figure (gl), node D
has finished removing the busy token and is in the repeat state again.

Therefore, frames are not repeated in the token holding state and the 1--kun sending state. In such a token passing system, the problem is how each node 1' actually controls access.

Figure 2 shows a state transition diagram. In order to distinguish between cases in which a token is captured and cases in which a token is not captured, a state in which no token is captured, called repeat state (1), is first defined. In this state, it only has the function of receiving frame data from transmitting nodes within the ring.

Additionally, in this state, it has a so-called repeat function that allows the same data to be streamed at the same time it is received. In this state, when a token is captured, the token is held.■
and is ready to send. When the frame data has been transmitted, the token is released, and the period until this state is reached is called the token sending state (■). After the token transmission is finished, the repeat state returns to ■. In this way, if the nodes in the ring are normal, each node will repeat the states ①, ②, and ②.

Each node also has the function of removing its own frame when it is not in a repeat state. Namely.

When you are in the 1-kun holding state■ and send out frame data, that frame J is sent all the way to you, and when the data comes back, that frame data is removed and the data goes around the ring many times. This includes a function to prevent this and confirm that frame absorption is completed.

In addition, each node has a function to regenerate tokens when they run out, but instead of randomly regenerating tokens, only nodes called Zunawara Active Monitors (AM) can regenerate tokens. ing. At this time, node 1, which is not an active monitor, ie, a normal node that does not perform token regeneration, is called a passive monitor (PM). In this system, furthermore, the state transition is a sequence for becoming an AM, a sequence for reproducing a token, or a sequence for issuing an abnormality notification frame when a ring is disconnected.

The state for reproducing a token is called the token regeneration state ■, the state in which PM transmits a frame to become an AM is called the monitor recovery state ■, and the state in which an error notification frame is being regenerated is called a beacon state This is called the sending state ■.

As a condition for becoming an AM, there is a timer that monitors the transition state, and these include timers 11 to 15. timer 1
1 is called the T1 timer, and it monitors the time that frames do not flow from the ring. Measure. This is a timer for token regeneration. The T2 timer 12 is activated in both PM and AM, but is a timer that monitors a relatively long time when only tokens do not flow. This means that, for example, in the case of an abnormal state such as a disconnection or no AM, the token will not flow. T2 time is usually 1
~2 seconds. When the T2 timeout occurs, the PM first enters the monitor recovery state (■) in order to determine that there is "no token" rather than a disconnection. The T3 timer 13 is a timer that limits the time that l--kun is held in the token holding state (2). The T4 timer 14 operates in the token sending state (2) and controls the timing of issuing tokens. In other words, after sending the frame and until the header is recognized, l・-
I monitor the time to make sure I don't get angry. That is, in the PM, when a frame is broken, a T4 timeout occurs, so it is determined that the frame is broken and a token is intentionally issued. The T5 timer 15 is a timer that measures the timing at which frames called monitor recovery frames are issued several times in the monitor recovery state (2).

Next, a sequence for generating an AM state using these timers T+-Ta will be explained. First, when the power is turned on, the node is in the PM repeat state.

In this state, if I-kun does not come, the token holding state (■) cannot be reached, so the Tt timer measures the time. When the T2 timeout (1 to 2 seconds) occurs, the node enters the monitor recovery state (■). A node in the monitor recovery state ■ sends out a frame for recovery called a monitor recovery frame. There are source and destination address fields 1' in the frame, and since the source address is its own address, this prevents conflicts between nodes in the ring that are repeating at the same time. In other words, a request to a node that wants to become an AM is to send a monitor recovery frame, but when it is in the monitor recovery state and receives a monitor recovery frame from another node, the source address (SA ) and my address (M
Compare with A). If S A >MA, the node abandons the monitor recovery state (■) and enters the repeat state (■). The frame is then routed downstream. If you repeat this, only one node with the maximum address among the nodes that have reached T2 timeout will be in the repeat state.■
■ Maintain the monitor recovery status. And Ta
(number is) time monitor Hawaii state ■ Continue. That is, the node recognizes that it can enter the AM state only when it sends out the monitor frame and SA=MA in the No 1 of the last remaining monitor state (■). Then, the state becomes the AM state repeat state (3) and becomes the normal reception state, and this becomes the only AM state in the ring. When 1-kun cannot be captured and the TI timeout (number -113) occurs in the repeat state ■ of the AM state of the node "," the AM node enters the token regeneration state ■, and a specific message called a ring purge frame (RPF) is sent. A frame is sent out. This is a state in which a frame can be streamed even though it does not have a token. When this RPF is sent out, the AM node l-" enters the l--kun sending state (2). In other words, the ring is again in the "token present" state, and this AM node is in the normal three states ■.

Continue ■ and ■. PM also continues in the states of ■, ■, and ■. In addition, if there is a disconnection somewhere in the ring, a certain node will transition from the repeat state ■ to the monitoring state ■, but in this case, since the wire is disconnected, its own ALUS MA will disappear. , the state changes to the point where the retry-out H0 is issued, and when this is repeated about 100 times without success, the state becomes the beacon sending state (■). This is because the node 1' below the location where the communication path in the ring is disconnected is in this state, and there is a possibility that the disconnection occurs between this node 1' and the node I of the L.

It is also possible to detect wire breaks.

Next, the three states mentioned above, namely repeat-like p,
Described below is a dedicated software that uses the l.-kun hold state and the l.-kun send state to perform normal control without involving processor control.

A token passing controller (TPC) 10 is included in each node adapter, as shown in FIG. 4, and a microprocessor 20 . ROM 30 for storing initialization routines. The received data 6 is connected to the common lot 60 together with the RAM 40 for storing programs and the transmitting/receiving buffer 50, and is transferred bit serially from other nodes on the ring communication path.
Input 00 while synchronizing in the repeat state,
The received frame data is transferred to the main memory RAM via the receive buffer 50 without the control of two microprocessors.
After DMA transfer, the microprocessor 20 processes the data and stores the result in the RAM 40. Among the results, data to be transferred to other nodes is transferred to the transmit/receive buffer 50 by DMA via the common data bus 60 for transmission.

When a token is transferred from a lower node, the TPO controls the transition from the repeat state to the token holding state, and returns to the 5 token sending state when the sending data frame from the sending buffer 50 is finished being sent. Control the transition. Finally, by sending out a token and detecting the last byte of the frame, control is executed to shift to the repeat state again. This T
Due to the existence of PO, such state transition for access control can be performed without being controlled by the microprocessor 20. As shown in FIG. 5, this TPO transfers hide data 1010 obtained by serial-to-parallel conversion of bit serial data 600 received in the repeat state to the receiving circuit 102 under the control of the control circuit 103, and performs synchronous control. and transfers it to the reception buffer 50. After capturing 1.-kun, the frame data transferred from the transmitting sofa 50 is transferred to the transmitting circuit under the control of the control circuit 103 from the token holding state to the token sending state, and the byte is sent from the repeat circuit 101 under synchronous control. Parallel to serial conversion and output frame and 1-kun in bit serial format 601
Output from

Next, the TPC repeat circuit 101 and its peripheral circuits will be explained using FIG. 6.

600 data RXD transferred in bit serial
is first converted into bytes using the serial/parallel conversion circuit R3R10]2.

At this time, first, to check whether it is a frame, a specific FS pattern (010101010) is connected to the data line 60.
0, and the matching circuit 1013 compares the first pattern of the frame with the same FS pattern and determines whether they match.
It becomes thicker. A frame is a 9-bit byte, e.g.
Starting with the 9 bits of the pattern, MAC3fl1 on and MACS (21 are each 9 hithide, 1! & destination arse and source alres are 6byl-, mmanl and data respectively) is n
High 1. Then, the fill is configured such that check=r-1 is 4 hides, file enl''(Fl'':) code 1- is 1 hide, and finally status is 1 byte. First, it is serial-parallel converted in R2H to become a high-item pattern, and the Pino 1 pattern is compared with the FS pattern, and if they match, it is confirmed that it is the beginning of Freno, so the flip-flop F is used to start frame reception.
F] The RXFRM signal is sent to OI4.

Since the next l hide of the FS pattern is 9 bits later, the 4-bit counter 101 is reset at the beginning of the frame.
When RCM = 8, the repeat circuit knows that R3H is carrying out the next byte seven times. This sequence is then repeated within the frame. On the other hand, BFA buffer l01
7 is for temporarily holding each byte. This temporarily held hide data is transferred to the IJ receiving circuit and is input to the receiving sequence counter (R3CN) and the receiving sequence counter (
RA CN) is used to identify each field I' within the frame. Then, the former calculates the number of bytes in field 1, and thereby knows what high l is. That is, if the hide is an address part, a comparison can be performed with a previously prepared address, for example, SA and MA can be compared. Further, if = 1man 1' part, it can be used to connect to the sequence of comparing SA and MA AI in the monitoring state (2).

Two registers B connected to HFA buffer 1017
FE and BFO are /' (7 doors) for synchronization.

In other words, there is a difference between the self-synchronized clock obtained by synchronizing the received data with the clock that is transmitted again, and to cover this, there are UFO and BFE, and each has a lead and a lag in the clock phase. The data is shifted and buffered according to the data, and bytes odd and even are set alternately. The frame input in this way is manually sent to the BFE and BFO, and the control circuit decides whether to take it out from there or not and registers the transmission register TSR.
Enter 1020. TSR1020 is a parallel-in/serial-out shift register.

The TFG flip-flops connected to this TSR1020 are 1. ! :It is a generator that generates O alternately.

When there is no frame to send, the TSR is 1010...
I am trying to send this pattern.

By doing this, it becomes possible to control whether the state is in a repeat state or not.

Next, correspond to each state in the state transition diagram of FIG.

The state flip-flop for state expression existing in the status control section of the TPO will be explained using FIG.

The TPC state flip-flop (F/F) has 5 active monitors (AM) and repeats (REP).

1-kun hold (TKNIID), token send (
T2l-KNPS), token regeneration (TKNGR), monitor recovery (MREC), beacon transmission (BCP)
FO and Fl each express each state of S).

F2. There are R3, R4, R5, and R6. First, when the power is turned on, the Fl repeat F/F is set to 1''.Alternatively, the Fl repeat F/F is set to 1" at the time of initial setting by firmware.
is set to "1".

At this time, this node enters the repeat state of the passing monitor state. When a token is input to the TPO repeat circuit 101, it is composed of three bytes: FS, MAC3, and FIE. This token is in turn BFA buffer 1017
When TCP recognizes token capture, F
The repeat F/F of 1 is set to 1", and the 1-kun holding F/F of F2 is set to 1".At this time,
The l--kun is in a state that has been absorbed by this node, and the token is not repeated, becomes a kill state, and is not forwarded to lower nodes. While the token holding F/F is set to “1”, the transmission buffer 5
The transmission frame prepared in step 0 is sent to TSR] 020 in the repeat circuit 101 for transmission.

At the end of transmission, it enters the l-kun sending state, resets the token holding F/F of F2, and sends the token of F3 III F
/ Set F to “1”. If you send a token after sending a frame, it will go into a repeat state again, so at the same time you reset the token sending F/F.

Set the FL repeat F/F to "1" again.

When the repeat F/F of Fl is set to "1" and the node is in the repeat state, when a time error occurs in the T2 timer, the monitor recovery F/F of the FS, which represents the monitor recovery state, becomes "l". Set. Then, the monitor frame is transferred to the ring via the TSR1020.

It checks whether the frame has been returned after circulating around the ring, and compares the source address (SA) in the frame with its own address (MA). As mentioned above, SA
> When MA, reset the FS monitor recovery F/F to return to the repeat state. And S.A.
= MA, this node becomes the active monitor.

At the same time as resetting the monitor F/F of F5, the active monitor F/F of FO is set to "11", and the F/F of Fl is also set to "l".In other words, this node is set to the active monitor F/F. The state becomes a repeat state.If it is possible to repeat 1-kun holding, token sending, and repeat, then F2.
The three states are repeated while F3 is set. However, in the repeat state, when the TI timer reaches T+timeout, the token regeneration F/F of F4 is set to "1'. When the token is regenerated and transferred to the ring, F4
The token is reproduced by setting the token sending F/F representing the token sending state to "1". Note that when comparing SA and MA in the passive monitor state, if SA≠MA is always within Ta time, this NO] cannot become an active monitor and will be in the beacon sending state, so the beacon sending F of F6 /F is set to 1".F6 is reset because it is in a repeat state when BCF is received, but it is set again when a disconnection is detected.

Next, the TPC MAC transition interrupt method will be explained using FIG.

TPC for each node 1ζ to perform MAC control.

In the state transition diagram shown in FIG. 2, the normal states (■, ■, ■) and the recovery states (■, ■, ■) are classified, and transitions between the normal states are performed by TPC without notifying the processor 20 of the node. is executed, and the transition between the repeat states is executed by the processor 20 as an "interrupt".

The feature is that the reason for the transition is presented to the firmware of the processor 20 by an interrupt cause display register. That is, the normal state control of frames and token transmission/reproduction formants and timing is automatically and independently performed by the TPO LSI without the firmware being aware of it.

Recovery for "no token" or "no AM" etc. is controlled by notifying the firmware by interrupt, thereby eliminating the two-hour loss. If all the state transitions in all 5 are executed by firmware, it will capture the token and send out the frame.
- There is a considerable delay in the transition of the normal state until sending out the kun.

During this time, other tasks that should be executed by the microprogram are not only put in a completely waiting state, but also the software load becomes extremely heavy. However, the control of accidental recovery is notified to the firmware and all execution is performed by the firmware, increasing efficiency. In other words, “
In the event of an accidental state such as "no token", "5", no AM, or "disconnection", the firmware itself prepares a frame for recovery, the firmware sends that frame consciously, and the resulting frame is also notified to the frameware. For example, when a monitor recovery frame is sent in the monitor recovery state 5, as a result, the comparison between SA and MA is based on the TPO hardware, that is, the LSI
However, the signal “become AM” is notified to the firmware by an interrupt. And the firmware is AM
If the firmware recognizes that it has become, the firmware itself detects the state. As described above, there are the following interrupts to the firmware in the recovery state.

T1 timer interrupt: The T1 timer operates in the active monitor repeat state (■), and if no token or frame passes, a timeout occurs in a few ms. This is notified to the firmware via an interrupt. The firmware changes the state to the token regeneration state (2) and sends out a Seso I-1, RPF (ring purge frame). When a token or frame passes, a T1 timer starts.

T2 timer interrupt: The T2 timer operates in a state other than the monitor recovery state ■ or the beacon sending state ■, and times out in a few seconds if no token passes. This is notified to the firmware via an interrupt. The firmware sets the state to the monitor state.

Send MRF. When i--kun passes, it restarts.

T5 timer interrupt: The T3 timer operates in the monitor recovery state.
The timer becomes 1 in a few ms.

This is notified to the firmware via an interrupt.

The firmware does not change the state and sends the monitor frame again.

AMM interrupt: In the monitor recovery state (2), when an SA-MA monitor recovery frame is received, it is notified by an interrupt. The firmware displays the status as active monitor copy 1/status■
to cents.

MRC release interrupt: In the monitor recovery state ■, when a monitor recovery frame with SA>MA is received, an interrupt is notified. The firmware enters the passive monitor repeat state■.

BCPS release interrupt: In the beacon sending state ■, when BCF is received, an interrupt is sent. Set the firmware to passive monitor repeat state■.

When an RPF of SA≠MA is received in the active monitor state of the AM resolution release interrupter, the firmware that notifies the interrupt sends it to the passive monitor repeat state.

Therefore, in the state flip-flops FO to F6 in FIG. 7, the signals marked with * are sent by the firmware.

Next, FIG. 8 shows a part of the TPC receiving circuit that generates interrupts, including the above-mentioned timers TI, T2 and T5.
A comparison between A and MA is also performed. 1iIF ahead of received frame
When the s pattern is confirmed, the byte data is transferred to the receiving circuit and the 5 receive sequence counter (R3CN)10
20 and a received address counter (RACN) 1021 can be used to identify each field within the frame. For example, if the byte input to the BFA buffer 1017 is the address part SA, the counters 1020 and 1021 are set in the decoding circuit 1022 to match the own address (MA) prepared in advance selected from the node address register 1023 using the decoding signal. Comparisons can be performed in comparison circuit 1024. When SA=MA, “0” at initialization
``1'' is set in the flip-flop set to .

Then, when a specific bit of the control register from the processor 20 becomes valid, that is, the output of the AND circuit 1027 becomes 1'', and the reception status 1028
A bit indicating SA-MA is inserted therein and provided to the control circuit section 103.

Reception status register RX? ' The contents of S are shown in Figure 8 (bl), the 1st bit and the O bit are the recovery frame that OO does not contain, and 01 is the ring barge frame (R
PF), 10 indicates the monitor recovery frame (MRF>) and 11 indicates the beacon frame (BCF), OO indicates SA=MA, 01 indicates S in the 3rd bit itself and the 2nd bit.
A>MA, 10 represents SA<MA, and 11 represents frame abort. The 4th bit, Fe2B, is incremented by seven nodes when an FC3 error is detected in the own node.

RCRC of error detection circuit 102 in Figure 8+a+
With register and CRC calculation circuit 1030! CR against Chenokko-1 who was attacked by 3FA buffer 1017
This is done by comparing the result of the C calculation with a constant 1032 in a comparison circuit 1031. If there is a discrepancy, Furno and Ware will be notified of the situation. Also.

At the end of the frame, the comparator circuit 1033 compares the received E pattern with the predetermined FE pattern, and in the case of a mismatch, an interrupt is sent to the fur J and the wear. .

Next, using Figure 1, each node is
A ring status interrupt scheme of the present invention is described in which the ring state can be checked to check the state of connection to the ring.

Figure 1 (a+) is a system in which a plurality of No.1s (for example, A, +1. 2 is a block diagram showing how a transmission frame 500 is transmitted to a ring from a transmission/reception buffer 50 in No. 1'A, and received after going around the ring for the purpose of checking whether or not the transmission/reception buffer 50 is connected. , when no FA is in transmit mode 1,
Naturally, frames cannot be received, and conversely, frames cannot be transmitted when the reception mode is 1'. Therefore, a transmit frame 500 is sent from the transmit/receive buffer to check whether No. 1"A is attached to the ring, but if the TPCIO receives that frame 500, it is sent directly from the repeat circuit of FIG. BFA buffer 1 via parallel conversion register 1012
The signal is latched at 017 and input to the receiving circuit shown in FIG.

If the high 1 input to the BFA hash sofa 1017 is the source address section SA, the own address (MA) output from the no 1' address register 1023
are compared. As described above, the comparison result between SA and MA is expressed in the third and second bits of the reception status register RXFS, and if it is 00, then SA=M8. At this time, as shown in FIG.
It is input to the ND circuit 1038. flip flop 10
Reference numeral 36 is a memory element that can be stored by the processor 20 of the no-FA, and is used to store the mode F for checking the connection state of the own node to the ring.

Set when ring wrap test is enabled.

In addition, the flip-flop 1037 is a storage element that indicates whether the status is valid or not, and it takes a considerable amount of time after a specific byte of the received frame is set in the BFA buffer until the status information is set in the RXFSIO28. Status information is RXFS 10
This indicates that the value has been definitely set to 028.

Therefore, if the ring loopback interrupt control circuit shown in FIG. Both bits are 0. In other words, when it is SA-MA, ANr)
The output of circuit 12 & 103B becomes a logic 1 and ring fold interrupt flip-flop 1039 is set. Therefore, an interrupt is generated to the firmware of the processor 20 only when another node FA receives a frame whose source status (SA) is the own node 1- (MA), and the processor stores the contents in the register. It will be expressed.

In this way, the interrupt conditions of the ring state interrupt method of the present invention are 2.
The ring wrap test enable bit is logic 1 and SA-MA. When SA=MA, node A is connected to the ring, but the condition is that no FCS error occurs at this time. FCS error detection is performed using the RCRC register 10 in the TPC of each node.
29, the CR calculation circuit 1030, and the comparison circuit 1031, so when the 3 processors display the interrupt contents in the register, the bit that indicates that an FCS error has been detected in another node and the FCS error in the own node. This would be a problem if it included a bit indicating that it was detected.

[Invention glue J fruit]

In the present invention, in each node 1'', the connection status of node 2' to the ring is checked by sending a flag to the own node, so that the source node is the own node and the test is valid. When Pino 1-1 is in a valid state, the TPC-dedicated hardware executes control to interrupt the processor by closing it and causing an interrupt to the processor, without going through the processor (CPII) hardware. , CPtJ is T
While the PC is performing the ring check ("can perform other tasks, therefore.

Not only does it direct the overall throughput.

This has the effect that the air load called Far J is reduced and system design is therefore extremely easy.

[Brief explanation of drawings]

FIGS. 1A+ and 1B are block diagrams illustrating the ring state interrupt control method of the present invention. FIG. 2 is a state transition diagram of the ring state interrupt control method of the present invention. FIG. 3 fat to fgl are diagrams explaining the token passing method. FIG. 4 is a diagram showing the configuration of the adapter. FIG. 5 is a block diagram of TI)C. Figure 6 is a repeat circuit diagram of TI)C. FIG. 7 is a diagram explaining the status control section of the TPC. FIG. 8 fat is a block diagram of a TPC receiving circuit. FIG. 8 (bl is a diagram explaining the reception status register of the TPC reception circuit. A, B, C... No. 1 20... Processor. 50... Transmission/reception buffer. 500...Transmission frame.

Claims (4)

[Claims] (1) In a system in which multiple nodes communicate with each other using a ring-shaped transmission path, one node transmits at least a frame to its own node, and when the result is ready, it is allocated. A ring state interrupt method characterized in that the state of connection with a ring is checked by notifying the processor of the node. (2) The ring state interrupt method according to claim 1, in which the interrupt is enabled only when a frame whose source address is the own node is received. (3) The ring state interrupt system according to claim 1, wherein enabling and disabling of interrupts can be controlled by a storage element that can be controlled by a processor of a node. (4) As the contents reported by the interrupt, the FC is
The ring state interrupt according to claim 1, characterized in that the ring status interrupt includes a first bit indicating that an S error has been detected, and a second bit indicating that an FCS error has been detected in the own node. method.