JP2007043382A - Frame processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a frame processor capable of reducing loads on a processor which relates to processing of unnecessary messages. <P>SOLUTION: The frame processor comprises a message agreement comparator 2300 which uses a message that is received first as a reference message for comparison, and compares messages that are then received and classified with the reference message for agreement; and a redundant message decision part 2400 for deciding whether the agreed message, is the one that has been received or not. The redundant message decision part 2400 discards a message if an elapsed time since the reference message is received is within a decision reference value, and clears the reference message if the elapsed time exceeds the decision reference value. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a frame processing apparatus that performs frame transmission / reception processing such as Ethernet (registered trademark) through a network interface and performs control such as discarding, passing, or sending out a received frame by identifying the frame.

With the spread of Internet home appliances, it has become essential to implement a control protocol, such as UPnP, that secures connectivity while reducing user effort. However, the following issues are expected for implementation. .
-The amount of messages exchanged between devices increases by reducing the user's effort.
-Implementation of extra functions such as traffic control is limited in cost, so control frames are output in bursts.
-Due to cost constraints, high-performance processors cannot be used in internet home appliances, so this traffic processing load becomes a burden.

As a conventional technique, there is a technique in which a simple TCP / UDP process is provided between a MAC and an IP layer to discard a frame and reduce the processing load on the processor (see, for example, Patent Document 1). In this prior art, the following processing is performed by providing a simple TCP / UDP processing between the MAC and IP layers. First, in the case of TCP, the TCP port number with which simple TCP / UDP processing is currently communicating, the sequence number to be received next, the sequence number to be sent next by the local host, and the confirmation ACK sent to the other party Get a sequence number that is not. If the number is smaller than the next sequence number to be received, it is determined as retransmission data, the data is discarded, an ACK specifying the correct number to be received next is sent, and then the sequence to be sent by the host If the ACK number is smaller than the number, it is discarded. In the case of UDP, the processor acquires the destination address / port number that is the object of processing, and discards the non-target frame. In this way, in TCP and UDP communications, frames that need not be processed by the processor are discarded to reduce the processing load.
Japanese Patent Laid-Open No. 2001-333096 (page 3 to page 5, FIG. 1)

  According to the UPnP regulations, information related to the valid period is given to the control message used for device control, and UPnP-compatible devices repeatedly transmit control messages having the same contents at intervals based on this valid period. The control message is received by another UPnP compatible device, and the control information of the issued UPnP compatible device is recognized to be still valid.

  In addition, in order to prevent the control message from reaching other devices due to the state of the transmission path, when the control message is transmitted at intervals based on the valid period, the control message having the same content is continuously transmitted as a spare. It is prescribed. However, the backup control message is an unnecessary message for the device when the UPnP-compatible device normally receives the control message based on the valid period.

  Although there are few devices connected in the current home network, it has not become apparent, but if the number of connected devices increases due to the spread of Internet home appliances in the future, the devices will have to process a lot of the above unnecessary messages .

  Unnecessary messages flow as messages in the application layer above UDP such as HTTP. Therefore, in a normal configuration in which processing of the application layer is performed by the processor, whether the received Ethernet frame has already been received and an unnecessary message is included is determined by performing normal frame reception processing by the processor, It cannot be determined until the message is taken out and compared with the received message. That is, when the number of unnecessary messages increases, the processing load on the processor increases even though there is no need for processing.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a frame processing apparatus that can reduce the load on the processor related to the processing of unnecessary messages.

  A frame processing apparatus according to the present invention is a frame processing apparatus that determines whether a specific frame determined from among frames received by a communication apparatus is to be passed or discarded, and is extracted using frame header information and payload information. A classification means for classifying the control messages included in the message, and a message received for the first time for each classified message as a reference message for comparison, with respect to the messages received and classified thereafter, the reference message and Message match comparison means for performing a match comparison of the above, and redundant message determination means for determining whether the message is the same message that has already been received with respect to a message that matches the reference message, The redundant message determination means receives from the time when the reference message is received. A measuring means for measuring the elapsed time is provided, and if the elapsed reception time until receiving the message to be judged is within the range of the criterion value defined based on the time, the message has already been received. If it is determined that there is a message and then discarded, and the reception elapsed time exceeds the determination reference value, the reference message is cleared, and the measurement means newly measures the reception elapsed time from the time when the message is received. To do.

  According to the above configuration, since the frame including the received message is discarded, it is possible to reduce the load on the processor related to the processing of unnecessary messages.

  In the frame processing apparatus of the present invention, the determination reference value is set in advance from the outside.

  According to the above configuration, it is possible to reduce the load on the processor accompanying the processing of unnecessary messages.

  In the frame processing device of the present invention, the determination reference value is specified by a value calculated using message validity period information calculated from a message, or the validity period information, header information, and payload information. It is a value calculated using the message refresh interval information calculated based on the specified protocol.

  According to the above configuration, it is possible to calculate the determination reference value based on the protocol specification according to the control message content, and to reliably discard unnecessary messages by the redundant message determination means, thereby further reducing the processing load on the processor Can be made.

  Further, in the frame processing device of the present invention, the message match comparison unit compares the reference message and the comparison target message to be compared by the message match comparison unit in units of bits, octets, or a combination thereof. Alternatively, a comparison using a data conversion operation to a certain length such as a hash operation is performed to determine a match / mismatch.

  According to the above configuration, it is possible to make a determination based on the overall matching of messages, and to compare messages in a short period of time.

  Further, in the frame processing device of the present invention, the message match comparison unit rearranges the information element fields in the message using format information in which the format of each information element field in the message is defined prior to the comparison, The format of the reference message is matched with the format of the message to be compared.

  According to the above configuration, even if the order of the information elements in the message is changed, the matching comparison can be performed.

  In the frame processing apparatus of the present invention, the message match comparison unit deletes the space / tab character string included in each of the reference message and the comparison target message prior to the comparison.

  According to the above configuration, even if the number of spaces and tabs in the message is different, the matching comparison can be performed.

  Further, the frame processing apparatus of the present invention includes a detecting unit that detects header information and payload information of a received frame to be responded, a specifying unit that specifies a transmission frame corresponding to the received frame from a detection result of the detecting unit, And extraction means for extracting the specified transmission frame from the transmission frame stream, and insertion means for creating a duplicate frame of the extracted transmission frame and inserting it into the transmission frame stream.

  According to the above configuration, a response message corresponding to the received message can be transmitted, and the processor processing load for performing the response can be reduced.

  The frame processing apparatus according to the present invention further includes an adding unit that applies a frame interval based on a preset value when the inserting unit inserts the duplicated frame into the transmission frame stream.

  According to the above configuration, the interval between the duplicated transmission frame and other frames can be adjusted, and the processing load when other devices receive these transmission frames can be reduced.

  According to the present invention, since a frame including a received message is discarded, it is possible to reduce a load on a processor related to processing of an unnecessary message.

  Hereinafter, a frame processing apparatus according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

[Embodiment 1]
FIG. 1 is a diagram showing a configuration of a frame processing apparatus according to Embodiment 1 of the present invention. The frame processing apparatus includes a network interface function 1000, a frame passing / discarding / response function 1100, and a frame processing function 1200.

  The network interface function 1000 transmits and receives Ethernet frames from a connected network such as a LAN. The frame passing / discarding / response function 1100 discards unnecessary frames that have been received from the network. The frame processing function 1200 performs protocol processing, routing processing, and the like, and transmits them to the network interface 1000 and other network interfaces (not shown) as necessary.

  FIG. 2 is a diagram showing a configuration of a frame passing / discarding / response function of the frame processing apparatus according to Embodiment 1 of the present invention. The received / received frame 2000 from the network interface 1000 is passed / discarded by the passage / discard determination unit 2600, the header classification function determination non-target pass / discard notification 2130, the message type classification function non-target pass / discard notification 2230, and the pass / discard determination result 2510. Passing or discarding is determined using this information. Then, the result is transferred to the frame processing function 1200 or discarded by the frame passing / discard / response function 1100.

  Next, the processing of the frame passing / discarding / response function configured as described above will be described.

  First, information of the received frame 2000 is input to the header classification unit 2100, and a protocol is specified by performing classification based on the header based on information set in advance by this frame passing / discard / response function, Decide whether to pass or discard. If the message is not subject to message classification, it is notified to the pass / discard function 2600 by a header classification part non-target pass / discard notification 2130 and processing is performed based on the pass / discard decision result. On the other hand, when the message is to be classified, the instruction is given to the message type classification unit 2200 of the corresponding protocol using the classification instruction 2110, and the payload information 2120 is also delivered.

  Upon receiving the instruction, the message type classification unit 2200 determines which message type the message included in the payload information corresponds to based on information set in advance, and includes header information included in the message. More specifically, the presence / absence of message content match comparison and passage / discard are determined. If the message content is not subject to comparison, the message type classification unit non-target pass / discard notification 2230 notifies the pass / discard function 2600 and performs processing based on the pass / discard decision result. On the other hand, when the message content match comparison target is selected, an instruction is given to the corresponding message content match comparison units 2300 using the comparison instruction 2210, and the format information / message content / protocol specification 2220 is also delivered.

  Receiving the instruction, each message content match comparison unit 2300 performs a match comparison of the message content based on the reference message information in order to determine which message the various message content in the specific message type corresponds to. If there is a mismatch, the fact is notified to the pass / discard determination function 2500 using the mismatch notification 2310. On the other hand, if they match, the corresponding redundant message determination unit 2400 is notified using the match notification 2320 and the message content / protocol specification 2330 is delivered.

  Each redundant message determination unit 2400 that has received the notification determines whether a received message is a redundant message by using a determination reference value in order to determine which message is redundant from a plurality of messages having the same content. Do. If it is a redundant message, a discard / instruction is given to the passage / discard determination unit 2500 using a passage / discard instruction 2410. On the other hand, when the message is not a redundant message, a passage instruction is similarly given.

  The pass / discard determination unit 2500 performs pass / discard determination using the plurality of received pass / discard instructions 2410 and the mismatch notification 2500, and notifies the pass / discard unit 2600 of the result using the pass / discard determination result 2510. To do.

  The passing / discarding unit 2600 that has received the notification determines passage / discarding of the received frame by using both the header classification unit non-target determination / passing notification 2130 and the message type classification unit non-target determination / passing notification 2230. , Perform the corresponding process.

  Next, the operation flow of the above functions assuming UPnP will be described.

  As the overall operation, the meaning of the message from a certain device is the same, but there are a mixture of fields with different field description orders and those with unnecessary character strings inserted, and the message itself has the same contents. The NOTIFY message is sent by multicast, received by the own device, and the preliminary message issued together with the status notification message periodically sent is discarded.

  FIG. 3 is a diagram showing a frame format input from the network in the frame processing apparatus according to Embodiment 1 of the present invention. Frames to which Ethernet / IP / UDP / SSDP is applied in order from the lower layer. FIG. 4 is a diagram showing the traffic pattern of the frame input from the network and the contents of the payload information in the frame processing apparatus according to Embodiment 1 of the present invention.

  Here, a case where a frame of the format shown in FIG. 3 is input to the frame processing apparatus shown in FIG. 1 in the traffic pattern shown in FIG.

  The network interface 1000 has a function of recognizing a frame by detecting a preamble, receiving a broadcast address / multicast frame addressed to itself, calculating an FCS, and transferring a normal frame to the inside. Since the Ethernet destination address DA3000 shown in FIG. 3 indicates multicast, the input frame is received by the network interface function 1000, the preamble 3800 is removed, and if the FCS calculation result is correct, the FCS 3900 is set as a normal frame. After removal, the frame is transferred to the frame passing / discarding / response function 1100. The transferred frame is input to the header classification unit 2100 as a frame 2000 (see FIG. 2).

  FIG. 5 is a diagram showing an internal configuration of the header classification unit in the frame passing / discarding / response function of the frame processing apparatus according to Embodiment 1 of the present invention. In the classification table 5000, the source MAC address, destination MAC address, ether type, VLAN PID, protocol identifier, source IPv4 address, destination IPv4 address, source IPv6 address, destination IPv6 address, source as header information for each entry Parameters such as port numbers and destination port numbers and corresponding message type classification functions are set in advance.

  Here, as the parameter value of entry A, destination MAC address: 01: 00: 5E: 7F: FF: FA, source MAC address: MAC-A, ether type: 0x0800, VALN PID: none, protocol identifier: 0x11, transmission Source IPv4 address: IP-A, Destination IPv4 address: 239.255.255.250, Source IPv6 address: None, Destination IPv6 address: None, Source port number: PortA, Destination port number: 1900, Message type classification function: α (protocol : SSDP, source IPv4 address: IP-A).

  The input frame 2000 is compared with the header information and the classification table 5000 by the header classification function 5100. If there is a matching entry, the corresponding classification instruction and payload information 5200 to 5500 are output. If there is no matching entry, a non-judgment pass / discard instruction 5600 is output in accordance with a preset instruction. Based on the information, the pass / discard unit 2600 performs a pass or discard operation.

  In this case, since it matches the entry A and the corresponding message type classification function is α, the classification instruction α5200 is output to the message type classification unit 2200 (see FIG. 2), and the payload information of the input frame 2000 is also included in the payload information. It is output as α5300.

  FIG. 6 is a diagram showing an internal configuration of a message type classification unit in the frame passing / discarding / response function of the frame processing apparatus according to Embodiment 1 of the present invention. The message type classification unit includes a message classification function for each corresponding protocol such as SSDP and a table attached to the function.

  Here, the part related to the message classification function 6000 (for SSDP) will be described, but the operation is the same for other message classification functions.

  In the classification table 6020, parameters such as NOTIFY in UPnP and the request line of the M-SERCH method are preset in each entry as message identification information for each protocol. Assume that parameters such as format information, valid period, refresh interval, and number of repetitions are set in advance. Here, “NOTIFY * HTTP / 1.1” representing the NOTIFY method is set as entry A of the classification table 6020, and format information “NOTIFY \ r \ n, HOST: \ r \ n, CACHE” is set as entry A of the protocol specification information 6010. -CONTROL: \ r \ n, LOCATION: \ r \ n, NT: \ r \ n, NTS: \ r \ n, SERVER: \ r \ n, USN: \ r \ n, \ r \ n ", enabled It is assumed that the period: 120 [sec], the refresh interval: 1/4, the number of repetitions: 2, the valid period header: CACHE-CONTROL: max-age =, and the corresponding comparison instruction α-A6001 are set.

  The message classification function 6000 is activated according to the input classification instruction. A character string from the beginning to the first line end code “\ r \ n” is extracted from the input payload information, and this is used as message type identification information and compared with a preset classification table 6020. . If there is a matching entry, a comparison instruction for the corresponding message type set in the entry is output, and the corresponding message content matching comparison function 2300 (see FIG. 2) is activated. On the other hand, if there is no matching entry, an out-of-judgment notification is notified to the out-of-judgment determination function 6200, and an out-of-judgment pass / discard instruction 6201 is output in accordance with a preset instruction here. A pass / discard operation is performed in the pass / discard function 2600.

  Since the payload information 5300 in this case matches the entry A, the corresponding comparison instruction 6001 is output to the message content match comparison function 2300, and format information 6002, message content 6003, and protocol definition information 6004 are also output.

  FIG. 7 is a diagram showing an internal configuration of the message content match comparison unit in the frame passing / discarding / response function of the frame processing apparatus according to Embodiment 1 of the present invention. The message content match comparison unit includes a plurality of message content match comparison functions for each message type, and the plurality of message content match comparison functions operate for one comparison instruction. Here, the part related to the message content match comparison function 7000 will be described, but the operation is the same for other message content match comparison functions.

  In the reference message information, information of messages having the same message identification information such as “NOTIFY * HTTP / 1.1” classified in the past but having different contents in other fields is registered. Here, it is assumed that the content shown in the message 4120 in FIG. 4 is registered as the reference message information 7020.

  The message content match comparison function 7010 that has received the comparison instruction 6001 performs comparison with the reference message 7020 using the received format information 6002 and message content 6003, respectively.

  In the comparison, first, the reference message 7020 is searched from the beginning, and the character codes corresponding to the spaces and tabs are deleted, and the message end code “\ r \ n \ r \ n” is found. Do this until As a result, it is possible to prevent a situation where a mismatch determination is simply made due to a difference in the number of spaces or the number of tabs.

  Next, the format information 6002 is searched from the beginning, the character string from the line end code “\ r \ n” to the point where “\ r \ n” is found is obtained, and “HOST: I get. Using this as the first comparison criterion, search for the reference message after deleting the character code corresponding to the space and tab from the beginning, and search for the next character string starting from the line end code "\ r \ n" Get up to "\ r \ n" which is the end code and get "HOST: 239.255.255.250:1900". This is compared with the reference character string “HOST:” at the beginning.

  In this case, since they are the same character string, the same processing is performed for the third and subsequent lines, and this is repeated until the message end code “\ r \ n \ r \ n” is found in the format information. On the other hand, if they do not match, the character string up to the next “\ r \ n” in the reference message is acquired and compared, and if it matches, the character string is used as the second line and the lines in the message are swapped and formatted. The same processing is performed for the third and subsequent lines of information. If the message end code “\ r \ n \ r \ n” does not match until it is found, it is determined that the message is not to be compared, a mismatch notification 7002 is output, and the process ends.

  The above processing is performed in the same manner for the message contents after deleting the character code corresponding to the space and tab. As a result, it is possible to prevent a situation where a mismatch determination is simply made by changing the rows.

  The reference message that has been formatted by the above method and the message content are compared using a bit unit, an octet unit, or a combination thereof, a data conversion operation to a certain length such as a hash operation, and the like.

  When the contents of the messages 4120, 4220, 4320, and 4420 in FIG. 4 are input, they are all determined to be in agreement, a coincidence notification 7003 is output to the redundant message determination unit 2400, and the content 7001 of the reference message is also output. The

  If there is a mismatch, a mismatch notification 7002 is output to the comparison entry function 7200. The comparison entry function 7200 counts the results by the mismatch result collection function 7210. If the content match comparison results of all message types do not match, a mismatch notification 7201 is sent to the pass / discard judgment unit 2500 as not being a comparison target message. Output. The same information is output to the entry registration / update / clear function 7230. Upon receiving this, the entry registration / update / clear function 7230 uses the entry information holding function 7220 to search for a message type content match comparison function having an empty entry or an updatable entry, and registers / updates the received message content. The next time the same message content is received, it will be subject to matching comparison.

  FIG. 8 is a diagram showing an internal configuration of the redundant message determination unit in the frame passing / discarding / response function of the frame processing apparatus according to Embodiment 1 of the present invention. The redundant message determination unit includes a plurality of redundant message determination functions corresponding to the message content match comparison function, and one redundant message determination function corresponds to one message content match comparison function. Here, a portion related to the redundant message determination function 8000 will be described, but the operation is the same for other redundant message determination functions.

  The determination reference value extraction / calculation function 8010 extracts necessary information from the received reference message content 7001 based on the protocol specification information 6004 and calculates a determination reference value 8011 together with other specification information.

  Here, the reference message content is the message 4120, and the information that the refresh interval is 1/4 and the validity period header is CACHE-CONTROL: max-age = is obtained from the protocol specification information. Search and obtain the character string from the line end code “\ r \ n” to the point where “\ r \ n” is found. This is the first comparison target, and it is compared whether the leading part of this character string is “CACHE-CONTROL: max-age =”.

  If there is a match, the subsequent character string is obtained and used as the valid period, but if it does not match, the character string up to the next “\ r \ n” in the reference message is obtained and compared continuously. To obtain the validity period.

  When the above is applied to the message 4120, 120 is obtained as the effective period, and this is multiplied by the refresh interval: 1/4 to obtain 30 [sec] as the determination reference value 8011.

  Here, when a plurality of frames shown in FIG. 4 are received, a coincidence notification 7003 is output by the determination function in the previous stage when Packet A4100 is received. At this timing, the determination reference value comparison function 8020 compares the determination reference value 8011 (30 [sec]) with the output value of the timer 8030. Assuming that sufficient time is available before the reception of PacketA 4100, in this case, the timer value is larger than the determination reference value, so the determination reference value comparison function 8020 determines that this is a newly received message, and the passage instruction 8021 is displayed. The timer reset instruction 8023 is output to the timer 8030 to return the timer to the initial value. Subsequently, when PacketB 4200 is received, a match notification 7003 is also output by the preceding determination function. At this timing, the determination reference value comparison function 8020 compares the determination reference value 8011 (30 [sec]) with the output value of the timer 8030.

  Since the time interval between PacketA 4100 and PacketB 4200 is 4110 (0.02 [sec]), which is smaller than the criterion value, the criterion value comparison function 8020 determines that this has already been received and is an unnecessary redundant message, and outputs a discard instruction 8022. . Subsequently, when PacketC 4300 is received, a match notification 7003 is output by the determination function in the previous stage. At this timing, the determination reference value comparison function 8020 compares the determination reference value 8011 (30 [sec]) with the output value of the timer 8030. Since the time interval between PacketA 4100 and PacketC 4300 is 4110 (0.02 [sec]) + 4210 (31 [sec]), which is larger than the criterion value, the criterion value comparison function 8020 indicates that this is a newly received message. Determination is made, a passage instruction 8021 is output, a timer reset instruction 8023 is output to the timer 8030, and the timer is returned to the initial value.

  Subsequently, when PacketD 4400 is received, a match notification 7003 is also output by the preceding determination function. At this timing, the determination reference value comparison function 8020 compares the determination reference value 8011 (30 [sec]) with the output value of the timer 8030. Since the time interval between PacketC 4300 and PacketD 4400 is 4310 (0.02 [sec]), which is smaller than the criterion value, the criterion value comparison function 8020 determines that this has already been received and is an unnecessary redundant message, and issues a discard instruction 8022. Output. The redundant message is determined by the above method, and the result is output.

  In this embodiment, an example in which the determination reference value extraction function 8010 calculates the determination reference value using the protocol definition information is shown. However, this value may be set using the external setting 8012.

  FIG. 9 is a diagram showing an internal configuration of the pass / discard determination unit in the frame pass / discard / response function of the frame processing apparatus according to Embodiment 1 of the present invention. The passing / discarding determination unit includes a passing / discarding determining function for each message type, and one passing / discarding determining function corresponds to each message type. Here, the pass / discard determination function 9000 will be described, but the operation is the same for other pass / discard determination functions.

  The pass / discard determination function 9000 is activated by a comparison instruction 6001. When the discrepancy notification 7201 is received, it is determined that the content of the message is not subject to comparison, and a determination result 9005 is output as “pass” because it is not possible to determine whether the message is a redundant message. When any of the passage instructions 8021 to 8121 is received, it is determined that the message is a newly received message in the message having the same message type and content, and the determination result 9005 is output as passing. When any one of the discard instructions 8022 to 8122 is received, it is determined that the message is already received in a frame having the same message type and content, and is an unnecessary redundant message, and the determination result 9005 is output as discard.

  The final determination function 9200 that has received this outputs the result to the pass / discard unit 2600 as the pass / discard determination result 9201. The passing / discarding unit 2600 performs actual frame passing / discarding control based on this information.

  As described above, according to the frame processing apparatus of the first embodiment, unnecessary messages are represented in traffic in which redundant messages are already received even though the message contents are the same, as represented by UPnP NOTIFY method. This can be determined and removed, and the load on the frame processing function can be reduced. Also, status notification messages periodically sent by UPnP compatible devices connected to the home network can be processed, and unnecessary messages sent as spare status notification messages can be discarded. In addition, it is possible to reduce the load on the processor accompanying unnecessary message processing while maintaining the latest state of each UPnP device.

[Embodiment 2]
Since the frame processing apparatus according to the second embodiment has a form in which a function is added to that of the first embodiment, the same reference numerals are given to the components already described in the first embodiment in the following description. Shall be omitted.

  Similar to the frame processing apparatus of the first embodiment, the frame processing apparatus of the second embodiment includes a network interface function 1000, a frame passing / discarding / response function 1100, and a frame processing function 1200.

  FIG. 10 is a diagram illustrating a configuration of a frame passing / discarding / response function of the frame processing device according to the second embodiment of the present invention. Only the components for the transmission frame will be described. First, the components 11100 to 111000 in the transmission frame stop operating until the “discard” instruction is output in the passage discard instruction 2410 in the component for the received frame. Always pass 11000.

  When the “discard” instruction is output in the communication discard instruction 2410, the frame to be inserted is transmitted from the frame processing function 1200 corresponding to the received frame having the same content that has passed immediately before. For this reason, this information is notified to the start / stop function 111000, and each of the components 11100 to 11900 is started by this function. The start / stop function 111000 holds an appropriate value in advance or a value set from the outside, and stops the function of each component as if there is no frame to be inserted after a certain period of time has elapsed since the start of each component Let When the insertion completion notification 11610 is output, the insertion target frame is found, and the function of each component is stopped assuming that the insertion process is completed.

  The header classification information creation function 11700 obtains, from the reception header classification unit 2100, header information 11710 related to the reception frame 2000 that is the target of discarding after activation. Based on this information, transmission header information corresponding to the received frame is created, and the header classification unit 11100 is set to determine a frame having this header as an insertion target.

  The message type classification information creation function 11800 obtains from the format information 2220 output by the received message type classification unit 2200 the format information 11810 related to the received frame 2000 that is to be discarded after activation. Based on this information, transmission message type information corresponding to the received message is created, and the message type classification unit 11200 is set to determine a frame having this message type as an insertion target.

  The message content classification information creation function 11900 obtains the message content 11910 related to the received frame 2000 to be discarded after activation from the message content 2220 output by the received message type classification unit 2200. Based on this information, transmission message content information corresponding to the received message is created, and the message content match comparison unit 11300 is set to determine a frame having this message content as an insertion target.

  It is determined whether the transmission frame 11000 is a response message to a frame received before the reception frame 2000 that has already been received and is discarded as an unnecessary redundant message. Therefore, the information of the transmission frame 11000 transferred from the frame processing function is input to the header classification unit 11100. Then, based on the information set by the header classification information creation function 11700, this function performs classification based on the header, determines whether the frame has the target header, and determines the presence / absence and passage of the message classification. When the message is not subject to message classification, the passage / insertion unit 11600 is notified by a header classification part determination non-target passage notification 11130 to perform the passage process. On the other hand, if the message is to be classified, the instruction is given to the message type classification unit 11200 of the corresponding protocol using the classification instruction 11110, and the payload information 11120 is delivered.

  Upon receiving the instruction, the message type classification unit 11200 determines which message type the message included in the payload information corresponds to based on the information set by the message type classification information creation function 11800. It is specified by the header information included in the message, and the presence / absence and passage of the message content match comparison is determined. If it is excluded from the message content match comparison target, the message type classification part determination target non-target pass notification 11230 notifies the pass / insert function 11600 and performs processing. On the other hand, when the message content match comparison target is selected, the message content match comparison unit 11300 is instructed using the comparison instruction 11210, and the format information / message content / protocol specification 11220 is delivered.

  Upon receiving the instruction, the message content match comparison unit 11300 determines to which message various message contents in a specific message type correspond. For this reason, the message content match comparison unit 11300 performs a match comparison of the message content based on the reference message information set by the message type classification information creation function 11900, and displays a match or mismatch determination result as a match notification 11320 or a mismatch notification 11310. Is used to notify the passage / insertion determination function 11500.

  Upon receipt of the notification, the pass / insertion determination unit 11500 performs pass / insertion determination using the received match notification 11320 and mismatch notification 11310, and uses the pass / insertion determination result 11510 to pass the result to the pass / insertion unit 11600. Notice.

  Upon receipt of the notification, the passing / inserting unit 11600 determines the passing or insertion of the received frame by using both the header classification unit non-target determination notification 11130 and the message type classification unit non-target determination notification 11230. When passing, the received frame is passed as it is, and when inserted, the received frame is copied and transmitted, and the processing corresponding to the discarded received frame is completed, and the insertion completion notice 11610 is sent to the start / stop function 111000. It outputs, stops each component, and waits until the received frame is discarded again.

  Next, the operation flow of the above functions assuming UPnP will be described.

  FIG. 11 is a diagram showing a frame format input from the network in the frame processing apparatus according to Embodiment 2 of the present invention. Frames to which Ethernet / IP / UDP / SSDP is applied in order from the lower layer. FIG. 12 is a diagram showing the contents of the traffic pattern and payload information of a frame input from the network in the frame processing apparatus according to the second embodiment of the present invention. The payload information is payload information 3700 (PacketA / B) and payload information 12700 (PacketC / D).

  Here, a case where the received frame shown in FIG. 11 is input to the frame processing apparatus of the second embodiment with the traffic pattern shown in FIG. 12 will be described.

  Details of the processing at the time of reception have been described in the first embodiment and will be omitted. As a result, PacketB 13200 shown in FIG. 12 is discarded as a redundant message.

  This discard is output as a pass / discard instruction 2410 to the start / stop function 111000 (see FIG. 10), and each function is started.

  As shown in the format 3000 to 3600 in FIG. 11, the header classification information creation function 11700 uses the header classification unit 2100 as the header information 11710 as the parameter value of the entry A corresponding to PacketB to be discarded. 00: 5E: 7F: FF: FA, source MAC address: MAC-A, ether type: 0x0800, VALN PID: none, protocol identifier: 0x11, source IPv4 address: IP-A, destination IPv4 address: 239.255.255.250 , Source IPv6 address: None, Destination IPv6 address: None, Source port number: PortA, Destination port number: 1900, Message type classification function: α (Protocol: SSDP, Source IPv4 address: IP-A) Thus, the classification information of the corresponding transmission frame is created.

  Specifically, the transmission source of each parameter relating to the address is set as the transmission destination, the parameters of the own apparatus held in advance as the transmission source are set, and the transmission source and the transmission destination are switched for the port number.

  In this way, as shown in 12000 to 12600 in FIG. 11, as transmission frame classification information, destination MAC address: MAC-A, transmission source MAC address: MAC-B, ether type: 0x0800, VALN PID: none, protocol Identifier: 0x11, Source IPv4 address: IP-B, Destination IPv4 address: IP-A, Source IPv6 address: None, Destination IPv6 address: None, Source port number: 1900, Destination port number: PortA, Message type classification Function: Acquires α (protocol: SSDP, source IPv4 address: IP-B).

  FIG. 13 is a diagram showing an internal configuration of the header classification unit in the frame passing / discarding / response function of the frame processing apparatus according to Embodiment 2 of the present invention. Information created by the header classification information creation function 11700 is set in the classification table 14000 as external setting information 14400. The frame 11000 input based on this information is compared with the classification table 14000. If they match, a classification instruction 14200 and payload information 14300 are output. If there is no matching entry, a non-judgment passage instruction is output. 14600 is output. This information is notified to the passage / insertion unit 11600 as a header classification unit non-target passage notification 11130, and a passage operation is performed with this function.

  In this case, since they match, the classification instruction 14200 is output to the message type classification unit 11200 (see FIG. 10), and the payload information of the input frame 11000 is also output as the payload information 11120.

  The message type classification information creation function 11800 uses the format information 11810 “M-SEARCH \ r \ n, Host: \ r \ n, Man: \ r corresponding to the packet B to be discarded as format information from the message type classification unit 2200. \ n, MX: 3 \ r \ n, \ r \ n ”. In this information, the character string from the beginning to the first line end code “\ r \ n” is extracted, and “M-SERCH” is acquired.

  Based on this information, a response message type and format information based on preset message correspondence information based on the protocol is created. Specifically, since “HTTP / 1.1 200 OK” corresponds to “M-SERCH” according to the protocol, it is set as message type identification information as external setting information 15400 in the classification table 15020. In addition, the format specified in the protocol specification as the M-SERCH response as the format information "HTTP / 1.1 200 OK \ r \ n, CACHE-CONTROL: \ r \ n, Location: \ r \ n, SERVER: \ r \ n, ST: \ r \ n, EXT: \ r \ n, USN: \ r \ n, \ r \ n "are set as the external setting information 15400 in the classification table 15020.

  FIG. 14 is a diagram showing an internal configuration of the message type classification unit in the frame passing / discarding / response function of the frame processing apparatus according to Embodiment 2 of the present invention. Information created by the message type classification information creation function 11800 (see FIG. 10) is set in the classification table 15020 as external setting information 15400.

  The message classification function 15000 is activated according to the input classification instruction. A character string from the beginning to the first line end code “\ r \ n” is extracted from the payload information input together. Using this as message type identification information, a comparison with a preset classification table 15020 is performed. If they match, a comparison instruction is output and the corresponding message match comparison means 11300 (see FIG. 10) is activated. . On the other hand, if there is no matching entry, a non-judgment pass instruction 15201 is output, and this information is sent to the pass / insertion unit 11600 as a message type classification part non-judgment pass notice 11230 (see FIG. 10). Perform the passing operation at.

  In this case, the payload information 14300 matches, a comparison instruction 15001 is output to the message match comparison means 11300, and format information 15002 and message content 15003 are also output.

  The message content classification information creation function 11900 (see FIG. 10) acquires the message content 11910 corresponding to PacketB that is the target of discard from the message type classification unit 2200 as the message content. The specific contents are as shown in the message 13220 of FIG.

  Based on this information, a reference response message based on preset message correspondence information based on the protocol and information on the device itself is created. Specifically, the response message format information output from the message type classification part “HTTP / 1.1 200 OK \ r \ n, CACHE-CONTROL: \ r \ n, Location: \ r \ n, SERVER: \ r \ n, ST: \ r \ n, EXT: \ r \ n, USN: \ r \ n, \ r \ n ", discarded message content" M-SEARCH * HTTP / 1.1 \ r \ n , Host: 239.255.255.250: 1900 \ r \ n, ST: urn: schemas-upnp-org: device: InternetGatewayDevice: 1 \ r \ n, Man: "ssdp: discover" \ r \ n, MX: 3 \ r Necessary information is extracted from “\ n, \ r \ n” and information set in advance by the own device, and each parameter is inserted.

  First, the information “max-age = 120” preset in the own device is inserted in “CACHE-CONTROL:”, and the information “IGD-HTTP” preset in the own device is also inserted in “Location:”. /1.1 UPnP / 1.0 UPnP-Device-Host / 1.0 '', insert `` urn: schemas-upnp-org: device: InternetGatewayDevice: 1 '' extracted from the message content into `` ST: '', and `` EXT: ”Is inserted from the information“ ”(null) preset in the own device, and“ USN: ”is extracted from the information“ uuid: 0208591c-00dc0301 :: ”preset in the own device and the message contents Insert "urn: schemas-upnp-org: device: InternetGatewayDevice: 1". This is set as the external setting information 16202 in the reference message information 16020.

  FIG. 15 is a diagram showing an internal configuration of the message match comparison unit in the frame passing / discarding / response function of the frame processing apparatus according to the second embodiment of the present invention.

  The message content match comparison function 16010 that has received the comparison instruction 15001 uses the received format information 15002 and message content 15003 to compare with the reference message 16020, respectively.

  In the comparison, first, the reference message 16020 is searched from the beginning, and the character codes corresponding to the spaces and tabs are deleted, and the message end code “\ r \ n \ r \ n” is found. Do this until This prevents a mismatch determination result due to a difference in space and the number of tabs. The same operation is performed for the message content 15003.

  Next, the format information 15002 is searched from the top, the character string from the line end code “\ r \ n” to the point where “\ r \ n” is found is obtained, and “CACHE- CONTROL: ”. Using this as the first comparison criterion, search for the reference message after deleting the character code corresponding to the space and tab from the beginning, and search for the next character string starting from the line end code "\ r \ n" Get up to "\ r \ n" which is the exit code and get "CACHE-CONTROL: max-age = 120". This is compared with the reference character string “CACHE-CONTROL:” at the beginning.

  In this case, since they are the same character string, the same processing is performed for the third and subsequent lines, and this is repeated until the message end code “\ r \ n \ r \ n” is found in the format information. On the other hand, if they do not match, the character string up to the next “\ r \ n” in the reference message is acquired and compared, and if it matches, the character string is used as the second line and the lines in the message are swapped and formatted. The same processing is performed for the third and subsequent lines of information. If the message end code “\ r \ n \ r \ n” does not match until it is found, it is determined that the message is not subject to comparison, and a mismatch notification 16201 is output to end the process.

  This is similarly performed for the message contents after deletion of the character codes corresponding to the spaces and tabs. As a result, it is possible to prevent a situation where a mismatch determination is simply made by changing the rows.

  The reference message that has been formatted by the above method and the message content are compared using a bit unit, an octet unit, or a combination thereof, a data conversion operation to a certain length such as a hash operation, and the like.

  If the message content of 13400, which is PacketC, is input, it is determined to be a match, and a match notification 16003 is output to the pass / insertion determination unit 11500.

  Based on the determination result, the pass / insertion determination unit 11500 notifies the pass / insertion determination result 11510 of the transmission frame 11000. The passing / inserting unit 11600 that has received the notification passes the transmission frame 11000 without doing anything in the case of a passing instruction, creates a copy of PacketC 13400 and inserts it as PacketD 13300 as shown in FIG. An insertion completion notice 11610 is output to instruct the start / stop function 111000 to stop (see FIG. 10). Note that, at the time of this insertion operation, the packet D 13300 may be sent out after a predetermined time based on preset information.

  Upon receiving the notification, the start / stop function 111000 stops each component and waits until the next received frame discard occurs.

  As described above, according to the frame processing apparatus of the second embodiment, it is possible to send a response message corresponding to a discarded message determined as an unnecessary redundant message that has already been received. The distance from the frame can be adjusted.

  The present invention discards a frame including a received message, and thus has an effect of reducing the load on the processor related to processing of an unnecessary message. In an apparatus for transmitting and receiving a frame such as Ethernet through a network interface, The present invention is useful for a frame processing apparatus that identifies a frame and performs control such as discarding, passing, or sending out a received frame.

The figure which shows the structure of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the structure of the frame passage / discard / response function of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the frame format input from a network with the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the content of the traffic pattern and payload information of the frame input from a network with the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the internal structure of the header classification | category part in the frame passage / discard / response function of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the internal structure of the message classification part in a frame passage / discard / response function of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the internal structure of the message content coincidence comparison part in a frame passage / discard / response function of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the internal structure of the redundant message determination part in a frame passage / discard / response function of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the internal structure of the passage / discard judgment part in a frame passage / discard / response function of the frame processing apparatus in Embodiment 1 of this invention. The figure which shows the structure of the frame passage / discard / response function of the frame processing apparatus in Embodiment 2 of this invention. The figure which shows the frame format input from a network with the frame processing apparatus in Embodiment 2 of this invention. The figure which shows the content of the traffic pattern and payload information of the frame input from a network with the frame processing apparatus in Embodiment 2 of this invention. The figure which shows the internal structure of the header classification | category part in the frame passage / discard / response function of the frame processing apparatus in Embodiment 2 of this invention. The figure which shows the internal structure of the message classification | category part in the frame passage / discard / response function of the frame processing apparatus in Embodiment 2 of this invention. The figure which shows the internal structure of the message matching comparison part in a frame passage / discard / response function of the frame processing apparatus in Embodiment 2 of this invention.

Explanation of symbols

1000 Network interface function 1100 Frame passing / discarding / response function 1200 Frame processing function 2100 Header classification unit 2200 Message type classification unit 2300 Message match comparison unit 2400 Redundant message determination unit 2500 Pass / discard determination unit 2600 Pass / discard unit 11100 Header classification unit 11200 Message type classification unit 11300 Message match comparison unit 11500 Pass / insertion determination unit 11600 Pass / insertion unit

Claims (8)

A frame processing device that determines passage or discard of a specific frame determined from frames received by a communication device,
Classification means for classifying control messages included in the frame, extracted using the header information and payload information of the frame,
A message match comparison means for performing a match comparison with the reference message for messages classified after received as a reference message that is a reference for comparison for the first received message for each classified message;
Redundant message determination means for determining whether the message is the same message that has already been received with respect to a message that matches the reference message,
The redundant message determination unit includes a measurement unit that measures an elapsed reception time from the time when the reference message is received, and the determination that the elapsed reception time until the message to be determined is received is defined based on the time. When it is within the range of the reference value, it is determined that the message has already been received and discarded, and when the reception elapsed time exceeds the determination reference value, the reference message is cleared, A frame processing apparatus for newly measuring the elapsed reception time from the time when the measuring means receives the message.   2. The frame processing device according to claim 1, wherein the determination reference value is set in advance from the outside.   The determination reference value is a value calculated using message validity period information calculated from the message, or a message calculated based on the protocol specified by the validity period information, header information, and payload information. 2. The frame processing device according to claim 1, wherein the value is calculated using the refresh interval information.   The message match comparison unit compares the reference message and the comparison target message to be compared by the message match comparison unit by a bit unit, an octet unit, or a combination thereof, or a fixed length such as a hash operation. The frame processing apparatus according to claim 1, wherein a comparison using a data conversion operation is performed to determine a match / mismatch.   Prior to the comparison, the message match comparison means rearranges the information element fields in the message using format information in which the format of each information element field in the message is defined, and the format of the reference message and the comparison target message 5. The frame processing apparatus according to claim 4, wherein the frame processing apparatus is matched with the format of the frame.   5. The frame processing apparatus according to claim 4, wherein the message match comparison unit deletes a space tab character string included in each of the reference message and the comparison target message prior to the comparison. Detecting means for detecting header information and payload information of a received frame to be responded;
Specifying means for specifying a transmission frame corresponding to the received frame from the detection result of the detection means;
Extraction means for extracting the specified transmission frame from the transmission frame stream;
Insertion means for creating a duplicate frame of the extracted transmission frame and inserting it into the transmission frame stream;
The frame processing apparatus according to claim 1, comprising:   The frame processing apparatus according to claim 7, wherein the inserting unit includes a adding unit that adds a frame interval based on a preset value when the duplicate frame is inserted into the transmission frame stream.

Images