WO2001025902A1 - Prozessorsystem, insbesondere ein prozessorsystem für kommunikationseinrichtungen - Google Patents
Prozessorsystem, insbesondere ein prozessorsystem für kommunikationseinrichtungen Download PDFInfo
- Publication number
- WO2001025902A1 WO2001025902A1 PCT/EP2000/009741 EP0009741W WO0125902A1 WO 2001025902 A1 WO2001025902 A1 WO 2001025902A1 EP 0009741 W EP0009741 W EP 0009741W WO 0125902 A1 WO0125902 A1 WO 0125902A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- command
- processor system
- execution unit
- execution
- data
- Prior art date
Links
- 230000015654 memory Effects 0.000 claims abstract description 19
- 239000000872 buffer Substances 0.000 claims description 26
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000012432 intermediate storage Methods 0.000 claims description 2
- 230000003139 buffering effect Effects 0.000 claims 1
- 238000003860 storage Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
- G06F9/3885—Concurrent instruction execution, e.g. pipeline or look ahead using a plurality of independent parallel functional units
Definitions
- processor system in particular a processor system for communication devices
- Communication controllers are used in m communication terminals, routers or gateways, which for the sake of simplicity are referred to below as the host system.
- the reception and transmission of communication information is generally managed or controlled.
- the communication information present in the form of digital data is fetched from a buffer and then, possibly packaged with further information bits, a transmission buffer is written.
- the content of the transmit buffer is then passed on to a modulator or transceiver, which modulates the digital transmit data onto a carrier signal, converts it to D / A and sends it to a receiver via a communication channel.
- the processing of communication information takes place in accordance with so-called communication protocols, which are constructed in the form of layers called "layers".
- layers layers
- the input data of the respective layer must be packed in each layer with a so-called protocol header and passed on to the subordinate layer.
- the protocol header must be taken from the respective input data and the data passed on to the respective higher-level layer.
- Header information from a received bit stream adding the header information to one to be sent Bitstream and the transfer of the respective data to another layer (ie the data transfer).
- a typical communication controller architecture is, for example, m "A 16Mb / s adapter chip for the IBM Token-Rmg Local Area
- US Pat. No. 5,434,976 proposes a communication controller which has two independent central processor units (CPUs), each processor unit having its own path for fetching or reading out a command to be executed and for decoding and executing the command read out.
- CPUs central processor units
- MAC layer Medium Access Control
- the other processor unit essentially executes host commands and manages the buffer memory functions associated with the reception or transmission of data.
- this architecture has the disadvantage that two separate processor units are used, which increase the required chip area and the power consumption.
- the present invention is therefore based on the object of proposing a processor system which on the one hand has a small chip area requirement and power consumption and on the other hand enables the functions required for processing communication information to be carried out effectively and quickly.
- the processor system according to the invention in particular in the form of a communication controller, only comprises a central processor unit (CPU) for executing instructions stored in a program memory, the processor unit comprising a single path for reading an instruction from the program memory and decoding the read instruction. Furthermore, a plurality of execution paths that can be operated in parallel are provided for the parallel execution of different commands or program sequences, each of which access the path used jointly for reading out and decoding a command.
- CPU central processor unit
- execution path It is advantageous if only one execution path has the normal computing and addressing functions of the CPU perceives, while another execution path only executes certain special functions, so that this execution path can be implemented more easily and the implementation effort can be further reduced. It is particularly advantageous if the last-mentioned execution path can only carry out a function that is required particularly frequently, and this can be, for example, moving data blocks, ie data transfer, when processing communication protocols.
- the command set of the processor system can advantageously be expanded in such a way that the corresponding "block move" command simultaneously specifies a programmable offset value with which the data block is written, for example, in an output port or read out from an output port.
- a bit offset can in particular be used as the offset value.
- Transfer rates can also be performed less frequently required or slower functions on the data bus with the lower transfer rate, while functions that are required more frequently or can be carried out quickly, such as in particular data transfers, can be carried out via the data bus with the higher transfer rate.
- This exemplary embodiment according to the invention thus differs from the known solutions, which either suggest the use of a single data bus in combination with a single processor unit or the use of two separate data buses in combination with two separate processor units.
- fast transfers of the so-called payload data can be carried out in this way on the data bus provided with a sufficient bandwidth and which has the higher transmission rate. while the protocol header data is processed via the slower data bus.
- FIG. 1 shows a simplified block diagram of a central processor unit or central unit (CPU) according to an exemplary embodiment of the present invention
- FIG. 2 shows the connection of the processor unit shown in FIG. 1 to a processor system according to the invention
- 3A-3C show enlarged representations of the input / output ports and registers shown in FIG.
- the processor unit 1 serving as a communication controller for a communication terminal, for example, comprises a path for reading out and decoding a command from a program memory 8 of the processor system, which in FIG. 1 in the form of function blocks 2 and 3 is shown.
- a parallel command execution is provided, wherein the parallelism is given only after the function blocks 2 and 3 and is formed by several execution units 5, 7 which can be operated in parallel.
- two such parallel execution units 5, 7 are provided for the independent processing of various commands or program sequences, with each execution unit being assigned a buffer memory or register 4 or 6 after decoding stage 3.
- These buffer memories 4, 6 each serve to store information which the subsequent execution unit 5 or 7 requires for executing the command.
- the execution units 5 and 7 are advantageously constructed or structured differently. To reduce the implementation effort, one of these execution units should be constructed as simply as possible so that only a limited number of functions or commands can be carried out, while at least one other execution unit can be kept as generally as desired and in particular can carry out all possible functions or commands should. In the illustrated
- the execution unit 7 is even simplified to such an extent that it is designed only for the execution of a specific command, preferably the command for moving a data block, while the execution unit 5 executes the entire command set of
- Processor system can run. In this way, the structure of the buffer 6 and the execution unit 7 can be compared the structure of the buffer 4 and the execution unit 5 can be simplified.
- the buffers 4 and 6 each serve to temporarily store information which is necessary for the execution of the command to be executed in the respective execution path.
- the information required in each case is stored in the buffers 4 and 6, in order in this way the path with the function blocks 2 and 3 for the parallel execution to release another command. If the execution unit 7 is only used to execute the command to move a data block, only one bit has to be in the corresponding buffer 6
- the execution path of the execution unit 5 can be carried out in parallel with another
- Program sequence can be assigned.
- Each program flow is assigned its own state variables, which form the state space of the respective program flow, also referred to as the "context”.
- separate hardware such as program counters and registers for storing the respective status variables (zero bits, carry bits, address pointer etc.)
- a command set is required for the so-called multitasking or parallel operation, which includes, among other things, commands for starting and stopping a program run, for setting a program run-specific priority or for waiting for a signal for program run synchronization.
- the execution unit 7 is preferably only used to carry out a data block move command, while the execution unit 5 can be designed as generally as possible in order to be able to execute all possible functions or commands.
- the execution unit 7 can thus carry out Data transfers are used while the execution unit 5 takes over the remaining protocol processing in parallel. Since the data transfers have to run relatively quickly compared to the protocol processing, it is advantageous to connect the execution unit 7 to a data bus with a relatively high transmission rate. This will be explained in more detail below with reference to FIG. 2.
- the central processor unit 1 is within the central processor unit 1
- the processor unit 1 can only access the data bus 9 with its execution unit 7, while the generally constructed execution unit 5 can use both data buses 5, 7 ,
- Input and output ports 11 and buffers or registers 12 are connected to the fast data bus 9 in order to implement the fast data transfers, in particular to shift the so-called payload data of communication protocols which have the actual communication information.
- the ports 11 are implemented as more complex special registers and serve as an interface to a transceiver 14 of the corresponding communication terminal, also referred to as a modulator, via which data are sent or received.
- An interrupt controller 16 evaluates, in particular, interrupt signals generated by the ports 11 and, depending on this, controls the central processor unit 1 in accordance with a predetermined interrupt handling.
- the buffers 12 form the interface to the respective host system 15 (ie the respective communication terminal, router or gateway), which writes or reads out the buffers 12 from the bit stream m to be transmitted or received.
- the buffers 12, which are implemented, for example, as RAM memory, can be controlled via DMA controllers that can be configured by the processor unit 1, so that the data transfers between the buffers 12 and the host system 15 do not impose any significant load on them 11
- FIG. 3B shows the signals applied to a port 11 with parallel data input and serial data output, this port differing from the port shown in FIG. 3A only in that n-bits are read in parallel on the input side and the output data are output serially.
- FIG. 3C also shows the signals applied to a buffer or a register 12, data words preferably being written and read out in the form of bytes in the buffers 12.
- an address signal ADR is applied, which in each case denotes the address of the buffer 12 which is to be accessed.
- the slower data bus 10 is connected to the actual data memory 13 of the processor system, which is designed in the form of a RAM mass memory and primarily for the intermediate storage of control information and
- Header information of a communication protocol is used.
- the components 11, 12 provided for the processing of communication protocols for the transfer of the so-called payload data are separated from the section provided for the processing of the protocol header data, and the execution unit 7 and the fast data bus 9 can be used for the quick payload data transfers are processed, while the processing of the header data of the respective communication protocol can run in parallel over the slower data bus 10 (and the embodiment 5).
- FIG. 3A shows the signals fed to a port 11 with serial data input and parallel data output.
- This port 11 receives input data D_IN serially, which are written to port 11 in accordance with a clock signal CLK n.
- the port 11 receives a 3-bit control information BP, which denotes the bit position of the data to be written, and a further 3-bit control information BW, which denotes the width or length of the bit field to be written.
- BP 3-bit control information
- BW which denotes the width or length of the bit field to be written.
- a data block of Lange BW is written in port 11 with a bit offset of Lange BP relating to the first bit position of port 11.
- A is used as a further control signal
- the data are read out in parallel with n bits, the reading of the data being enabled via a further control signal D_READ.
- an interrupt signal D_READY is provided, which generates an interrupt when data is stored in port 11. 11
- FIG. 3B shows the signals applied to a port 11 with parallel data input and serial data output, this port differing from the port shown in FIG. 3A only in that n-bits are read in parallel on the input side and the output data are output serially.
- the signals applied to a buffer or a register 12 are also shown in FIG. 3C, data words in the form of bytes preferably being written and read into the buffers 12.
- an address signal ADR is applied, which in each case denotes the address of the buffer 12 which is to be accessed.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Communication Control (AREA)
- Advance Control (AREA)
- Executing Machine-Instructions (AREA)
- Multi Processors (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/089,907 US7185184B1 (en) | 1999-10-06 | 2000-10-05 | Processor system, especially a processor system for communications devices |
EP00966119A EP1224540B1 (de) | 1999-10-06 | 2000-10-05 | Prozessorsystem, insbesondere ein prozessorsystem für kommunikationseinrichtungen |
DE50002014T DE50002014D1 (de) | 1999-10-06 | 2000-10-05 | Prozessorsystem, insbesondere ein prozessorsystem für kommunikationseinrichtungen |
JP2001528795A JP3651672B2 (ja) | 1999-10-06 | 2000-10-05 | 演算処理システム、特に通信装置のための演算処理システム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE19948099.0 | 1999-10-06 | ||
DE19948099A DE19948099A1 (de) | 1999-10-06 | 1999-10-06 | Prozessorsystem, insbesondere ein Prozessorsystem für Kommunikationseinrichtungen |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2001025902A1 true WO2001025902A1 (de) | 2001-04-12 |
Family
ID=7924660
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2000/009741 WO2001025902A1 (de) | 1999-10-06 | 2000-10-05 | Prozessorsystem, insbesondere ein prozessorsystem für kommunikationseinrichtungen |
Country Status (6)
Country | Link |
---|---|
US (1) | US7185184B1 (de) |
EP (1) | EP1224540B1 (de) |
JP (1) | JP3651672B2 (de) |
KR (1) | KR100471516B1 (de) |
DE (2) | DE19948099A1 (de) |
WO (1) | WO2001025902A1 (de) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW561740B (en) * | 2002-06-06 | 2003-11-11 | Via Tech Inc | Network connecting device and data packet transferring method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5179530A (en) * | 1989-11-03 | 1993-01-12 | Zoran Corporation | Architecture for integrated concurrent vector signal processor |
EP0529913A2 (de) * | 1991-08-26 | 1993-03-03 | International Business Machines Corporation | Datenverarbeitungssystem mit mehreren Ausführungseinheiten |
EP0767425A2 (de) * | 1995-10-06 | 1997-04-09 | Digital Equipment Corporation | Register und Befehlssteuerung für einen superskalaren Prozessor |
US5655124A (en) * | 1992-03-31 | 1997-08-05 | Seiko Epson Corporation | Selective power-down for high performance CPU/system |
US5954811A (en) * | 1996-01-25 | 1999-09-21 | Analog Devices, Inc. | Digital signal processor architecture |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SE397013B (sv) * | 1976-12-17 | 1977-10-10 | Ellemtel Utvecklings Ab | Sett och anordning for att overfora datainformationer till tva parallellt arbetande datamaskindelar |
US5027317A (en) * | 1989-03-17 | 1991-06-25 | Allen-Bradley Company, Inc. | Method and circuit for limiting access to a RAM program memory |
US5146558A (en) * | 1990-01-19 | 1992-09-08 | Canon Kabushiki Kaisha | Data processing system and apparatus |
JPH0520062A (ja) | 1991-07-09 | 1993-01-29 | Fuji Xerox Co Ltd | データ処理装置 |
EP0606299B1 (de) * | 1991-10-04 | 2003-09-10 | Bay Networks, Inc. | Verfahren und vorrichtung für simultane paketbus. |
JPH07502358A (ja) * | 1991-12-23 | 1995-03-09 | インテル・コーポレーション | マイクロプロセッサーのクロックに依るマルチプル・アクセスのためのインターリーブ・キャッシュ |
US5392437A (en) * | 1992-11-06 | 1995-02-21 | Intel Corporation | Method and apparatus for independently stopping and restarting functional units |
US5598362A (en) * | 1994-12-22 | 1997-01-28 | Motorola Inc. | Apparatus and method for performing both 24 bit and 16 bit arithmetic |
US5841771A (en) * | 1995-07-07 | 1998-11-24 | Northern Telecom Limited | Telecommunications switch apparatus and method for time switching |
US5732251A (en) * | 1996-05-06 | 1998-03-24 | Advanced Micro Devices | DSP with register file and multi-function instruction sequencer for vector processing by MACU |
-
1999
- 1999-10-06 DE DE19948099A patent/DE19948099A1/de not_active Withdrawn
-
2000
- 2000-10-05 JP JP2001528795A patent/JP3651672B2/ja not_active Expired - Fee Related
- 2000-10-05 US US10/089,907 patent/US7185184B1/en not_active Expired - Fee Related
- 2000-10-05 KR KR10-2002-7004290A patent/KR100471516B1/ko not_active IP Right Cessation
- 2000-10-05 EP EP00966119A patent/EP1224540B1/de not_active Expired - Lifetime
- 2000-10-05 DE DE50002014T patent/DE50002014D1/de not_active Expired - Lifetime
- 2000-10-05 WO PCT/EP2000/009741 patent/WO2001025902A1/de active IP Right Grant
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5179530A (en) * | 1989-11-03 | 1993-01-12 | Zoran Corporation | Architecture for integrated concurrent vector signal processor |
EP0529913A2 (de) * | 1991-08-26 | 1993-03-03 | International Business Machines Corporation | Datenverarbeitungssystem mit mehreren Ausführungseinheiten |
US5655124A (en) * | 1992-03-31 | 1997-08-05 | Seiko Epson Corporation | Selective power-down for high performance CPU/system |
EP0767425A2 (de) * | 1995-10-06 | 1997-04-09 | Digital Equipment Corporation | Register und Befehlssteuerung für einen superskalaren Prozessor |
US5954811A (en) * | 1996-01-25 | 1999-09-21 | Analog Devices, Inc. | Digital signal processor architecture |
Also Published As
Publication number | Publication date |
---|---|
JP3651672B2 (ja) | 2005-05-25 |
DE50002014D1 (de) | 2003-06-05 |
DE19948099A1 (de) | 2001-04-19 |
US7185184B1 (en) | 2007-02-27 |
JP2003511753A (ja) | 2003-03-25 |
EP1224540B1 (de) | 2003-05-02 |
KR20020035628A (ko) | 2002-05-11 |
EP1224540A1 (de) | 2002-07-24 |
KR100471516B1 (ko) | 2005-03-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE3751514T2 (de) | Adressieranordnung für RAM-Puffer-Steuereinrichtung. | |
DE3851111T2 (de) | Netzwerksystem mit Tokenprinzip-Zugriffsverfahren. | |
DE3751091T2 (de) | Übertragungsprotokoll zwischen Prozessoren. | |
DE69329607T2 (de) | Programmierte Ethernetanpassungseinrichtung mit frühzeitiger Unterbrechung für Beschleunigung von Datenübertragung | |
DE69131971T2 (de) | Netzwerkanpassungseinrichtung mit Puffer und Mehrbeschreibungskreis | |
DE69032384T2 (de) | Datenpuffer | |
DE69328320T2 (de) | Vorrichtung und Verfahren zur Datenübertragung zu und von einem Wirtrechnersystem | |
DE69323861T2 (de) | Multiprozessorsystem mit gemeinsamem Speicher | |
DE69519117T2 (de) | Ein Rechnersystem | |
DE3642324C2 (de) | Multiprozessoranlage mit Prozessor-Zugriffssteuerung | |
DE4121446C2 (de) | Terminal-Server-Architektur | |
EP0893755B1 (de) | Pufferspeicheranordnung | |
DE69032723T2 (de) | FIFO mit variablem Zuweisungszeiger | |
EP2030118B1 (de) | Mehrprozessor-gateway | |
DE68927795T2 (de) | Übertragungssystem mit einem gemeinsamen Speicher | |
EP1599794B1 (de) | Prozessor mit verschiedenartigen steuerwerken für gemeinsam genutzte ressourcen | |
DE69219390T2 (de) | ATM-Zellenvermittlungssystem | |
DE69121893T2 (de) | Koppelelement und verfahren zu seiner steuerung | |
DE69522900T2 (de) | System und Verfahren zur Verarbeitung von Befehlen ohne Abfrage der Hardware durch die Software | |
EP1230590B1 (de) | Prozessorsystem | |
DE69033416T2 (de) | Hauptspeicherkarten mit Einzelbit-Setz- und Rücksetz-Funktion | |
DE60131355T2 (de) | Paketverarbeitungsvorrichtung | |
EP1370952B1 (de) | Kommunikationsverfahren zur realisierung von ereigniskanälen in einem zeitgesteuerten kommunikationssystem | |
WO2001025902A1 (de) | Prozessorsystem, insbesondere ein prozessorsystem für kommunikationseinrichtungen | |
DE60022186T2 (de) | Unterhaltung einer entfernten Warteschlange unter Benutzung von zwei Zählern in der Verschiebesteuerung mit Hubs und Ports |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): JP KR US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2000966119 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020027004290 Country of ref document: KR |
|
ENP | Entry into the national phase |
Ref country code: JP Ref document number: 2001 528795 Kind code of ref document: A Format of ref document f/p: F |
|
WWP | Wipo information: published in national office |
Ref document number: 1020027004290 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10089907 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 2000966119 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 2000966119 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 1020027004290 Country of ref document: KR |