WO2005015386A1 - プロセッサ集積回路及びプロセッサ集積回路を用いた製品開発方法 - Google Patents
プロセッサ集積回路及びプロセッサ集積回路を用いた製品開発方法 Download PDFInfo
- Publication number
- WO2005015386A1 WO2005015386A1 PCT/JP2004/011661 JP2004011661W WO2005015386A1 WO 2005015386 A1 WO2005015386 A1 WO 2005015386A1 JP 2004011661 W JP2004011661 W JP 2004011661W WO 2005015386 A1 WO2005015386 A1 WO 2005015386A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- unit
- program
- computing
- computing unit
- integrated circuit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/20—Handling requests for interconnection or transfer for access to input/output bus
- G06F13/28—Handling requests for interconnection or transfer for access to input/output bus using burst mode transfer, e.g. direct memory access DMA, cycle steal
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7839—Architectures of general purpose stored program computers comprising a single central processing unit with memory
- G06F15/7842—Architectures of general purpose stored program computers comprising a single central processing unit with memory on one IC chip (single chip microcontrollers)
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Definitions
- the present invention relates to a processor integrated circuit and a product development method equipped with the processor integrated circuit, and in particular, a processor integrated circuit over a processor operated by a program and a product development method equipped with the processor integrated circuit. It is about Background art
- Figure 11 shows a block diagram of an audio product.
- the audio product is an audio processor LSI 900 that performs data compression processing and a control microcomputer that controls a single processor LSI 900. 01, a CD controller 902 for reading data from a CD, and a recording medium 903 for storing compressed data.
- the CD controller 1902 reads the music data from the CD and outputs an audio signal S 6202 to the processor LS 1900 together with the timing signal S 6201.
- Processor LSI 900 is a digital signal processor (hereinafter referred to as DSP) to be described later, compresses audio data signal S 6202 and records it on a recording medium 903.
- DSP digital signal processor
- FIG. 12 is a diagram showing an internal configuration of a conventional processor integrated circuit 900.
- the low speed computing unit 910 is capable of operation up to 50 MHz
- the high speed computing unit 920 is capable of 100 MHz operation.
- the high-speed calculator 920 enables 100 MHz operation by increasing the number of pipeline processing stages from the low-speed calculator 910, and the programs of the low-speed calculator 910 and the high-speed calculator 920 are not compatible with each other. .
- the low speed computing unit 910 is connected to the program memory 911 and the data memory 912, and performs compression recording processing by a dedicated program stored in the program memory 911.
- the processing result is stored in the data memory 912.
- the arithmetic unit 910, the program memory 911 and the data memory 912 are combined to form a low speed DSP 919.
- the low-speed DS P 919 and the processor LSI 900 external are all connected via the DMA controller 915.
- the DMA controller 915 arbitrates the internal bus access request from the control microcomputer 901, etc., the write request of audio data from the CD controller 902 by the timing signal S 6201, and the data read request from the recording medium 903 Perform DMA (Direct Memory Access) via 910.
- the high-speed DSP 929 is composed of a high-speed processor 920, a program memory 921 and a data memory 922, and is connected to an external LSI via a DMA controller 925.
- the DMA controller 925 also arbitrates access requests from outside the same LSI as the DMA controller 915.
- FIG. 13 is a DMA timing diagram of a conventional low speed DSP 919 and a DMA timing diagram of a conventional high speed DSP 929.
- the low-speed DSP 919 outputs the DMA read signal S 9100 at one clock after the DMA request, while the high-speed DSP 929 outputs the DMA read signal S at three clocks after the DMA request.
- Output 9200 That is, the latency of the low-speed DS 919 DMA is 1 clock, and the high-speed DS 909 is 3 clocks.
- the DMA controller 915 for low-speed DSP fetches the data on the next clock after the DMA request is issued, while the DMA controller for high-speed DSP-the 925 at the third clock after the DMA request is issued. Operates to capture data.
- the processor LSI 900 configured in this way is used to switch between two different DSPs according to the content of the target processing.
- the noise frequency of 50 MHz is required for single-speed recording processing.
- perform compression recording processing using a low-speed DSP 919 For example, it is assumed that the noise frequency of 50 MHz is required for single-speed recording processing.
- perform compression recording processing using a low-speed DSP 919 For example, it is assumed that the noise frequency of 50 MHz is required for single-speed recording processing.
- perform compression recording processing using a low-speed DSP 919 For example, it is assumed that the noise frequency of 50 MHz is required for single-speed recording processing.
- a clock frequency of 100 MHz is required.
- the low speed DSP 919 can not operate at 100 MHz. Therefore, compression recording processing is performed by the high speed DSP 929, the output selector 990 is switched, and the output of the high speed DSP 929 is recorded in the recording medium 903.
- the present invention solves the above-mentioned conventional problems, and provides a processor integrated circuit capable of ensuring program compatibility and speeding up without increasing the size of hardware and software, designing man-hours, and power consumption.
- the purpose is to Disclosure of the invention
- a processor integrated circuit is a computing unit group including two or more types of computing units, and a program for operating the computing units.
- a second storage unit that is a memory area used by the computing unit in the arithmetic processing, a single computing unit that performs arithmetic processing of the computing unit group, and A first connection switching unit for connecting to the first storage unit; a second connection switching unit for connecting one computing unit that performs arithmetic processing of the computing unit group to the second storage unit;
- the first and second storage units are shared to perform arithmetic processing by an arithmetic unit that constitutes the arithmetic unit group.
- a processor integrated circuit capable of reducing memory, achieving both compatibility of programs and speeding-up, without increasing circuit size and power consumption, and achieving both program compatibility and high speed. Can be reduced.
- a processor integrated circuit according to claim 2 of the present invention is the processor integrated circuit according to claim 1, wherein the operation unit group operates an operation unit operated with a program having a history of use. It is included.
- a processor integrated circuit according to claim 3 of the present invention is the processor integrated circuit according to claim 2, wherein the operation unit group operates with a program having a track record of use. It includes a computing unit with higher processing power. According to the present invention, the processor can be speeded up.
- a processor integrated circuit according to claim 4 of the present invention is the processor integrated circuit according to claim 3, wherein the computing unit operated by the program having a track record in use has the processing capability. It consumes less power than a tall computing unit. According to the present invention, it is possible to use a processor with low power consumption.
- a processor integrated circuit has a computing unit group including two or more types of computing units, and a plurality of memory areas, and at least one of the plurality of memory areas.
- a second storage unit storing one or more programs for operating the computing unit, and a plurality of memory areas used by the at least one computing unit to perform arithmetic processing
- a first connection switching unit for connecting a storage unit, an arithmetic unit performing arithmetic processing of the arithmetic unit group, and a memory area of the first storage unit storing a program used by the arithmetic unit
- a second connection switching unit respectively connecting the memory area of the second storage unit, and an arithmetic unit performing arithmetic processing of the arithmetic unit group, and a plurality of the arithmetic units are used.
- the first and second connection switching units control the connection between each computing unit and each memory area of the first and second storage units, thereby performing parallel arithmetic processing by each computing unit as the first and second connection switching units.
- the second storage unit is shared.
- the present invention it is possible to provide a processor integrated circuit capable of reducing memory, achieving both compatibility of program and speeding-up, without increasing circuit size and power consumption, and achieving both program compatibility and speeding up.
- a processor integrated circuit capable of reducing memory, achieving both compatibility of program and speeding-up, without increasing circuit size and power consumption, and achieving both program compatibility and speeding up.
- parallel arithmetic processing by the arithmetic unit of the above it is possible to provide a higher performance processor integrated circuit without increasing the circuit scale.
- a processor integrated circuit according to claim 6 of the present invention is the processor integrated circuit according to claim 5, wherein the first storage unit is a parallel arithmetic processing by a plurality of the operators.
- the first connection switching unit is configured to store a plurality of programs executed by each computing unit at the same time, and each memory of each computing unit and the first storage unit. By controlling the connection with the area, each arithmetic unit performing parallel arithmetic processing is connected with each memory area in which the plurality of programs are stored.
- a processor integrated circuit is characterized by: In the processor integrated circuit according to item 5, each arithmetic unit performing the parallel arithmetic processing starts input / output of data with the same control signal, and a ratio of processing units of the input / output data is The program period of each operation unit is used as a ratio to synchronize each operation unit.
- the processor integrated circuit according to claim 8 of the present invention arbitrates access to each processor with n (n is a natural number of 2 or more) processors having different DMA (Direct Memory Access) latency,
- n is a natural number of 2 or more processors having different DMA (Direct Memory Access) latency
- the clock frequency ratio of each processor is made the ratio of (DMA latency of each processor + 1) to perform arithmetic processing
- the arithmetic unit receives the first DMA request signal, it then invalidates (DMA latency 1) / 2 DMA request signals.
- each DMA arbitration circuit can be made identical, and only one type of control circuit around the processor can be provided, thereby reducing the number of hardware design steps.
- a product development method is a product development method performed using a processor integrated circuit having an arithmetic operation group consisting of two or more types of arithmetic operation units.
- First product development using the k-th arithmetic unit which is an arithmetic unit that operates with a predetermined program stored in the first storage unit that stores a program for operating the arithmetic unit.
- a program corresponding to the kth computing unit stored in the first storage unit the computing unit having a processing capability higher than that of the kth computing unit in the computing unit group.
- the second storage unit which is a used memory area, and the second storage unit
- the second connecting step of connecting with the X computing unit and the X-th computing unit are executed to execute the program, and the performance is improved compared with the product developed in the first product development step, or 1 Different functions from products developed in the product development process And a second product development step of developing a product to be possessed.
- the product developed in the second product development process can be developed without requiring a dedicated processor integrated circuit, and the increase in circuit scale can be further suppressed, and in the first product development process
- the process of changing from the program for the arithmetic unit used for the product to be developed to the program for the arithmetic unit used for the product to be developed in the second product development process may be performed simultaneously with the mass production process of the product developed in the first product development process. It is possible to hide the time required for changing the product developed in the second product development process to the program for the computing device.
- a product development method is the product development method performed using a processor integrated circuit having an arithmetic unit group including two or more types of arithmetic units.
- the first product development which develops a product using the k-th arithmetic unit which is an arithmetic unit which operates with the predetermined program stored in the first storage unit, which stores the program for operating the arithmetic unit.
- the second memory unit which is the memory area used Executing the program using the second connection step of connecting to the X operation unit, the power reduction step of reducing the power supply voltage of the processor integrated circuit, and the X operation unit; It has the third product development process for developing a product with the same performance or function as the product developed in the development process and with reduced power consumption.
- a product development method is a product development method performed using a processor integrated circuit having an arithmetic unit group including two or more types of arithmetic units.
- the kth operation that is an operation unit that operates with a predetermined program stored in the first storage unit that stores a program for operating the operation unit.
- a first product development step of developing a product using a scanner a program corresponding to the kth computing element stored in the first storage unit; at least the kth computing element and the kth computing element
- the program is executed using two or more computing units including the connection step, and the k-th computing unit and the X-th computing unit, and the performance is improved compared to the product developed in the first product development process. Or a function different from the product developed in the first product development process It is obtained by a fourth product development process to develop a product having.
- the product developed in the fourth product development process can be developed without requiring a dedicated processor integrated circuit, and the increase in circuit scale can be further suppressed, and in the first product development process
- the process of changing from the program for the computing unit used for the developed product to the program for the computing unit used for the product developed in the fourth product development process may be performed simultaneously with the mass production process for the product developed in the first product development process. It has the effect of being able to conceal the time required for changing to the program for the computing unit.
- a product development method is the product development method performed using a processor integrated circuit having an arithmetic unit group including two or more types of arithmetic units.
- the first product development which develops a product using the k-th arithmetic unit which is an arithmetic unit which operates with the predetermined program stored in the first storage unit, which stores the program for operating the arithmetic unit.
- a second storage unit which is a memory area used in The second connecting step of connecting to the Xth computing unit and the program executed using the Xth computing unit, and the performance is improved compared to the product developed in the first product development step, or Different from the product developed in the first product development process
- the product developed in the fourth product development process can be developed without requiring a dedicated processor integrated circuit, and the increase in circuit scale can be further suppressed, and in the second product development process
- the process of changing from the program for the computing unit used for the developed product to the program for the computing unit used for the product developed in the fourth product development process may be performed simultaneously with the mass production process for the product developed in the second product development process. It has the effect of being able to conceal the time required for changing to the program for the computing unit.
- FIG. 1 is a block diagram of a processor integrated circuit according to Embodiment 1 of the present invention.
- FIG. 2 is a block diagram of a processor integrated circuit according to Embodiment 2 of the present invention.
- FIG. 3 is a block diagram of a processor integrated circuit including peripheral circuits of low speed D SP and high speed D SP shown in FIG. 2;
- FIG. 4 is an operation timing diagram of low speed D SP and high speed D SP according to a second embodiment of the present invention.
- FIG. 5 is a diagram of the processor integrated circuit shown in FIG. 3 provided with a dividing circuit.
- FIG. 6 is a DMA timing chart of low speed D SP and high speed D SP according to a second embodiment of the present invention.
- FIG. 7 is a block diagram of a low speed DSP and a high speed DSP DM according to a second embodiment of the present invention.
- FIG. 8 is a block diagram of the low speed DSP and high speed DSP in the processor integrated circuit of FIG. The figure which shows the example provided 2 each.
- FIG. 9 is an audio product development flow diagram according to Embodiment 3 of the present invention.
- FIG. 10 is a diagram showing an operation mode in each product of DSP in the third embodiment of the present invention.
- Fig. 11 shows the configuration of the audio product.
- FIG. 12 is a block diagram of a conventional processor integrated circuit.
- Figure 13 is a DMA timing diagram of the conventional low speed DSP and high speed DSP.
- FIG. 1 is a diagram showing an internal configuration of a processor integrated circuit 100 according to a first embodiment of the present invention.
- the processor integrated circuit 100 includes a low speed computing unit 110 capable of operation up to 50 MHz, which is a computing unit group, and a high speed computing unit 120 capable of operating up to 100 MHz, a low speed computing unit 110 or a high speed computing unit
- a program memory 131 which is a first storage unit for storing a program for operating the 120, and a data memory, which is a memory area used by the low-speed calculator 110 or the high-speed calculator 120 during arithmetic processing.
- a low speed computing unit 110 which is a first connection switching section for connecting the low speed computing unit 110 which performs computing processing or the slow speed computing unit 120, and the program memory;
- the low speed operator 110 is a second connection switching unit that connects the data processing unit 110 or the high speed operation unit 120 to the data memory 132.
- DSP Digital signal Processor
- DMA controller 115 for low-speed DSP, which is a DMA arbitration circuit that performs DMA (Direct Memory Access), and when high-speed calculator 120 is used as high-speed DSP 129. It consists of a DMA controller 125 for high-speed DSP, which is a DMA arbitration circuit that performs DMA.
- the low-speed DSP 119 includes a low-speed calculator 110 and a first memory unit, a program memory.
- the high speed D SP 129 is composed of a high speed computing unit 120, a program memory 131 which is a first storage unit, and a second storage unit. It consists of a certain schedule memory 132.
- the processor integrated circuit 100 according to the first embodiment is a computing unit having a low-speed computing unit 110 that is a computing unit that operates with a program that has been used, and a computing unit having a processing capability higher than that of the low-speed computing unit 110 It is assumed that the low speed computing unit 110 consumes less power than the high speed computing unit 120.
- the processor integrated circuit shares the program memory 131 and the data memory 132 with the arithmetic processing by the low-speed calculator 110 and the high-speed calculator 120 having different processing power and power consumption. It is done.
- the low-speed operator 110 is connected to the program memory 131 and the data memory 132 by the operator selectors 141 and 142, and compression recording is performed by the dedicated program stored in the program memory 131.
- the processing result is stored in the data memory 132.
- the dedicated memory stored in the program memory 131 is connected to the program memory 131 and data memory 132 by the high-speed operator 120 force operator selectors 141 and 142.
- the program performs compression recording processing, and the processing result is stored in the data memory 132.
- the computing unit selector 141 which is the first connection switching unit switches the connection destination of the program memory 131 to the low-speed computing unit 110 or the high-speed computing unit 120 according to the program stored in the program memory 131.
- the computing unit selector 142 which is the second connection switching unit switches the connection destination of the data memory 132 to the low speed computing unit 110 or the high speed computing unit 120 operated by the program stored in the program memory 131.
- the low speed DSP 119 and the high speed DSP 129 are different from the conventional DSPs 919 and 929 in that the memory and computing unit are separated from the conventional DSPs 919 and 929 in FIG.
- DMA controller 115 controls my An internal bus access request from a controller (not shown), an audio data write request from a CD controller (not shown) by the timing signal S1201 and a request from a recording medium (not shown) Arbitrate read request S 1 3 0 0 and perform DMA via low-speed calculator 1 1 0.
- the high-speed D SP 1 2 9 is similarly connected to the L S I outside through the DMA controller 1 2 5.
- the DMA controller 125 also arbitrates access requests from the same LSI as the DMA controller 115.
- the processor integrated circuit according to the first embodiment of the present invention as described above shares the program memory 1 31 and data memory 1 32 with the arithmetic processing by the low-speed operator 1 1 0 and the high-speed operator 1 2 0.
- low-speed DSP 1 1 9 When used as low-speed DSP 1 9 9, store the low-speed calculator 1 1 0 program in program memory 1 3 1 and set the calculator selector 1 4 1 and 1 2 2 to the low-speed calculator
- the same operation as the operation unit can be performed in the same program as a program that has been used, and when it is used as a high-speed DSP 1 2 9 that has higher processing capacity than a low-speed DSP 1 1 9
- the high-speed operation can be performed by storing the program of the high-speed calculator in program memory 1 31 and switching the calculator selectors 1 4 1 and 1 2 2 to the high-speed calculator side.
- Can be halved Circuit scale power consumption without increasing the can you to provide a processor integrated circuit which can achieve both compatibility security and speed of the program.
- n is 2 or more
- the present invention is also effective when using a group of arithmetic units).
- n computing units may be used as computing units of the same type.
- a program for a low speed computing unit Although a certain program was used, it is not limited to this, and a new program may be used as a program for low-speed arithmetic units.
- FIG. 2 is a diagram showing an internal configuration of a processor integrated circuit 200 according to a second embodiment of the present invention.
- the first storage unit of the processor integrated circuit according to the second embodiment includes program memories 1311, 1312, 1313 and 1314 which are four memory areas.
- the program memory 131 which is the first storage unit is divided into four modules.
- the second storage unit includes data memories 1321, 1322, 1323, and 1324 which are four memory areas, and the data memory 132 which is the second storage unit in the first embodiment includes four data memories 132. It is divided into modules.
- Arithmetic unit selectors 1411, 1412, 1413, and 1414 which are the first connection switching units, each module so that the connection with the arithmetic unit that performs arithmetic processing can be switched for each of the four modules of the divided program memory. It is provided individually for each, and corresponds to the operation unit selector 141 which is the first connection switching unit in the first embodiment. Similarly, the second selector unit 1421, 1422, 1423, 1424 also switches the connection with the arithmetic unit that performs arithmetic processing to each of the four modules in the memory. It is provided individually for each module, and corresponds to the operation unit selector 142 which is the second connection switching unit in the first embodiment.
- the same operation as that of the first embodiment can be performed by switching all of the program memories 1311, 1312, 1313, 1314 or the data memories 1321, 1322, 1323, 1324 in conjunction with each other. Furthermore, by connecting the program memory 1311, 1312 and the data memory 1321, 1322 to the low speed computing unit 110, and connecting the program memory 1313, 1314, data memory 1323, 1324 to the high speed computing unit 120, a low speed DSP 219 It can perform parallel arithmetic processing as two processors of and high-speed DSP 229. For example, the compressed recording program described in the background art is divided into two programs of “audio compression processing” and “recording determination processing”, and the audio compression program is changed for the high-speed calculator 120.
- each program downloads the program of “recording judgment process” to the program memory 1311, 1312 corresponding to the low speed DSP 219, and the program memory of “voice compression process” corresponds to the high speed DSP 229. Down separately.
- the program of “recording judgment processing” and “voice compression processing” are downloaded together in program memory 1311, 1312, 1313.
- the program selectors 1411, 141 2, 1413, 1414 are used to connect each program memory storing the program of "recording judgment processing” with the low speed computing unit 110, and the program of "audio compression processing” is stored.
- Each program memory may be connected to the high speed computing unit 120. This will allow you to download the program once.
- the program memories 1311 and 1312 store a program of “recording determination processing” by digital watermark detection.
- the low-speed DSP 119 uses the data memory 1321 and 1322 in accordance with the program of "recording judgment processing” to detect "may I record” information embedded in the input audio data as a digital watermark. Do.
- the program memory 1313, 1314 stores a program of "audio compression processing".
- the high-speed DSP 229 compresses the input audio data using the data memory 1323, 1324 according to the "audio compression processing" program.
- the high-speed DSP 229 reads out the information detected by the low-speed DSP 219 and records the data compressed by the compression program on a recording medium (not shown) if it can be recorded.
- parallel processing of low-speed DSP 2 19 operating at 50 MHz and high-speed D SP 229 operating at 100 MHz provides a processing capacity equivalent to 150 MHz and triples the recording function. It can correspond to.
- FIG. 3 is a diagram showing a processor integrated circuit 300 including low speed DSP 219 and high speed DSP 229 peripheral circuits in FIG. 2, and the same reference numerals are used for the same components as in FIG. 1 and FIG. Use, I omit the explanation.
- the processor integrated circuit 300 is provided with clock gates 365, 366, 367 and 368, and the DMA controller for low speed DSP is set by the clock ON / OFF control circuit 361 from the control microcomputer (not shown). 1 15, low speed DSP 219, DMA controller for high speed DSP 125, high speed DSP
- the clock of the high speed DSP 229 and the high speed DSP DMA controller 125 is stopped.
- the clocks of the low speed D SP 219 and the low speed D SP DMA controller 115 are stopped.
- the power consumption does not increase when operating the conventional program using the low-speed computing unit 110.
- all the clocks are turned on.
- the power supply may be cut off instead of stopping the clock. It goes without saying that the same effect can be obtained by adopting the same configuration also in the processor integrated circuit of FIGS. 1 and 10.
- the timing signal gates 375 and 376 are the start control circuit.
- the DMA controller 115 for low-speed DSP and the DMA controller 125 for high-speed DSP are used as audio signals at both edges of the timing signal S 1201 which is the same control signal controlled by the setting signal gates 375 and 376. Takes data signal S 1202 and performs DMA write access to data memory in DSP.
- the recording function is "audio compression processing” and digital watermark It is divided into “recording judgment processing” by detection, and the input audio data fetched into the data memory is parallel-computed at low speed D SP 219 and high speed D SP 229.
- the low-speed DSP 219 has a frame period of 4 samples, detects digital watermark information contained in audio data for each frame, and outputs the detected information to the data memory 1321.
- the high-speed DSP 229 sets the frame period to 8 samples and compresses the audio data on the data memory every frame.
- the processor integrated circuit according to the second embodiment can set the ratio of program cycles of high speed DSP and low speed DSP by setting the ratio of processing units of low speed DSP 219 and high speed DSP to “1 to 2” samples. Synchronize each DSP as “1 to 2”.
- FIG. 4 is a timing diagram showing a synchronous relationship between low speed DSP 219 and high speed DSP 229 in a processor integrated circuit according to Embodiment 2 of the present invention.
- the control microcomputer (not shown) confirms that both DSPs are ready to start operation, and releases the timing signal gates 375 and 376 using the start control circuit 370 at time Tal.
- the low speed DSP 219 and the high speed D SP 229 start to load the audio signal S 1202 into the data memory by the DMA controller from time T a 1 by the same control signal.
- the low-speed DSP 219 starts detection processing of digital watermark information contained in frame 0, and starts updating old record judgment information on the data memory 1321 from time Tb2, information at time Tc1. Complete the D 1 update. Thereafter, the low speed D SP 219 repeats the above operation in four sample cycles.
- the high speed D SP 229 reads out the recording determination information D 1 detected by the low speed D SP 219. If the recording judgment information D1 is recordable, the high speed DSP 2 29 starts compression processing of the audio data of frame A, and the processing is completed by time Td l. After that, the high-speed DSP 229 repeats the above operation in an 8-sample cycle.
- the recording judgment information outputted by the low speed DSP 219 is a period during the updating, and the information is not decided.
- the frame start (time Tc 1 and Td 1) of high-speed DSP 229 is synchronized with the frame start of low-speed DSP 219. 229 does not read the data being updated at the beginning of each frame (time Tc l, Td 1).
- the two processor sets are synchronized by setting the ratio of the processing period of the low-speed DSP and the high-speed DSP to “1 to 2” samples, which is the ratio of the program cycles of the two DSPs.
- the processing unit for each DSP is “1 to 2”, the present invention is not limited thereto. “1 to N” samples are used as the program cycle ratio of each DSP. The present invention is effective even when used.
- FIG. 5 is a diagram of the processor integrated circuit described in FIG. 3 provided with a divide-by-two circuit 501.
- the divide-by-two circuit 501 supplies a clock S 2801 of 1 ⁇ 2 frequency of the input clock S 1802 to the low speed DSP DMA controller 115, the low speed DSP 219, and the high speed DSP DMA controller 125.
- the high speed DSP 229 is supplied with the clock S 1802 before division.
- the ratio “1: 2” of the clock supplied to the high speed DSP DMA controller 125 and the clock frequency supplied to the high speed DSP 229 is different from that of the conventional processor integrated circuit 900.
- FIG. 6 is a timing diagram showing DMA accesses of low-speed DSP 219 and high-speed DSP 229
- FIG. 7 is a pipeline operation of low-speed DSP 219 and high-speed DSP 229 when DMA requests are continuous. Is a timing diagram showing
- the low-speed DSP 219 and high-speed DSP 229 have different internal pipeline stages, as described in Figure 13, so the DMA latency of the low-speed DSP 219 is 1 clock, and the DMA latency of the high-speed DSP 229 is 3 clocks. is there.
- Second The DMA of the low speed DSP 219 shown in FIG. 6 is the same as the DMA of the low speed DSP shown in FIG.
- the DMA of the high-speed DSP 229 shown in FIG. 6 has a DMA request signal S 1250 for high-speed DSP of 2 clock widths of the clock S 1802 of the high-speed DSP 229. It is different.
- the DMA controller 125 outputs a DMA request signal from the DMA controller 125 for the high-speed DSP 229.
- the high-speed arithmetic unit 120 starts accepting DMA at the rising edge of the DMA request signal and accepts DMA every other clock. It is different. Then, data is read from the data memory in D SP three clocks after DMA is received.
- the clock frequency ratio of low-speed DSP and high-speed DSP is fixed at 1: 2, and the DMA controller of high-speed DSP is driven by clock S2801 with 1 Z 2 frequency of clock S 180 2
- the timing for reading data is the same time Tr as for the low speed DSP 219 and the high speed DSP 229.
- the DMA controller 115 and the DMA controller 125 can be made completely common circuits by making the DMA request acceptance of the high-speed DSP 229 effective every one clock.
- the DSP clock frequency ratio is set to “1: 2” for low-speed and high-speed DSPs whose DMA latency is “1 clock: 3 clocks”, when expansion is generally performed,
- the clock frequency ratio between the first processor and the second processor is “DMA latency of the first processor + 1: DMA latency of the second processor + 1”, and the arithmetic unit performing the arithmetic processing performs DMA reception “( It is good to ignore "(DMA latency-1)-2" out of DMA latency + 1) Z2.
- the processor integrated circuit according to the second embodiment of the present invention as described above includes: program memories 1311, 1312, 1313 and 1314, which are a plurality of memory areas; and data memories 1321, 13.22 and 1323, which are a plurality of memory areas. , 1324 and a selector that switches the connection with each memory area to a low-speed computing element or high-speed computing element Can reduce the memory by half, and can provide a processor integrated circuit that can achieve both program compatibility and high speed without increasing circuit size and power consumption. By combining this with low-speed and high-speed operators, parallel processing can be performed as a low-speed DSP and a high-speed DSP, and further, the processor integrated circuit can be speeded up.
- the low-speed DSP has a 4-sample cycle
- the high-speed DSP has an 8-sample cycle
- the frame start of the high-speed DSP and the frame start of the low-speed DSP are synchronized, eliminating the need for handshaking between the low-speed processor and the high-speed processor. it can.
- the clock frequency ratio between the high-speed DSP DMA controller 125 and the high-speed DSP 229 is “1: 2”
- the low-speed DSP 229 and the high-speed DSP 219 have the same data read timing
- the high-speed DSP 229 Since the DMA request acceptance of the above is made effective every one clock, the DMA controller 115 and the DMA controller 125 can be made to be a completely common circuit, and it is possible to prevent an increase in the number of designing steps of the DMA controller.
- n arithmetic operation unit groups different from each other is used, and n arithmetic operation unit groups may be the same arithmetic operation unit.
- a total of four low-speed operators and two high-speed operators 120A and 120B, and two low-speed operators, L channel and R channel, respectively.
- the four DSPs of low-speed DSP 219 A and 219 B and high-speed DSP 229 A and 229 B can perform four parallel arithmetic processing.
- processing of 300 MHz equivalent without additional memory You can gain the ability and get 6x speed.
- FIG. 9 is a development flow diagram of a new audio product using processor integrated circuit 500 of Embodiment 2 of the present invention.
- an existing processor integrated circuit 800 with only a low-speed DSP operating on a program with a proven performance is designed with a 0.1 micron pitch process.
- the processor integrated circuit 500 is the processor integrated circuit described in the second embodiment, and is designed with a 0.1 micron pitch process.
- the arithmetic unit group of processor integrated circuit 500 is a low-speed arithmetic unit 1 1 which is a k-th arithmetic unit operated by a program for operating processor integrated circuit 800. It is assumed that it is composed of a high-speed computing element 1 20 which is an X-th computing element having a processing capability higher than that of the 0-th and k-th computing elements.
- the existing audio product 10 using processor integrated circuit 800 is a portable model that performs single-speed audio recording using program P10 for low-speed DSP.
- the configuration of the existing product 10 is that in which the conventional processor LSI 900 of the product shown in FIG. 11 is a processor integrated circuit 800, and the present embodiment 3 relates to this.
- the product will be the product based on the development of four new products described later.
- Figure 10 shows the mode of use of low-speed DSP 2 1 9 and high-speed DSP 2 2 9 for each of the new products 1, 2, 3 and 4 in the flow chart of Fig. 9, supply voltage, operating frequency It is the figure which put together. This will be described below with reference to FIGS. 9 and 10.
- the first new product 11 which is a product developed in the first product development process, is a second generation portable model capable of recording at 1 ⁇ speed, and instead of the processor integrated circuit 800 in the existing product, It is a product whose cost has been reduced by replacing the processor integrated circuit 500.
- the program memory 1 3 1 1, 1 3 1 2, 1 3 1 3 1, which is the first storage unit having a plurality of memory areas described in FIG.
- Data of the second storage unit having a plurality of memory areas is connected to the low speed computing unit 110 by switching all of the 314 in the first switching unit, that is, the computing unit selectors 1411, 1412, 1 413, and 1414.
- the second new product12 a product developed in the second product development process, is a stationary model that can record at double speed (flow 0).
- the low speed DSP program P10 stored in the program memory 1311, 1312, 1313, 1314 of the processor integrated circuit in the first new product 11 It is assumed that a program P12 for high-speed DSP for operating the computing unit 120 is used.
- the first connection step all of the program memories 1311, 1312, 1313 and 1314 which are the first storage units in FIG. 2 are processed by the operation unit selectors 1411, 1412 and 1413 which are the first switching units. , 1414 are connected to the high speed computing unit 120, and in the second connection step, the data memory 1321, 1 which is the second storage unit in FIG.
- All of 322, 1323 and 1324 are a second switching unit, an arithmetic unit selector 1
- the process of changing the program P10 for low-speed DSP into the program P12 for high-speed DSP is the first This process can be performed simultaneously with the mass production process of new products11, and can hide the time required for program change. Also, newly
- the third new product13 a product developed in the third product development process, is a third-generation, low-power-consumption portable model that can record at 1x speed (flow 1).
- the third new product 13 is the same as the second new product as a professional. Further, the program memory 1311, 1312, 1313, 1314, and the data memory 1321, 1322, 1323, 1324 in FIG. The high-speed computing unit 120 is connected in the same manner as the second new product.
- the power consumption of LSI is proportional to the clock frequency and proportional to the square of the supply voltage. Also, if the power supply voltage is lowered, the wiring delay inside the LSI will increase and the normal LSI will not operate, but it can be operated if the clock frequency is lowered. Therefore, in the power reduction process, the power supply voltage is lowered by setting the clock frequency supplied to the high-speed D SP 329 used for the third new product 13 to 1 Z 2 of the processor integrated circuit in the second new product 12 Can.
- the clock frequency of the DSP is the same as that of the first new product 11, but by changing the voltage of LS 1200 to 1.5 to 1.2 V, the power consumption is 36% higher than that of the first new product 11. It is lowered. Also, the same high-speed DSP program P12 as the second new product 12 can be used, and there is no need to change the program.
- the fourth new product 14 that is a product developed in the fourth product development process is a stationary model that can record at triple speed (flow 2).
- the high speed DSP program P 12 stored in the program memories 1311, 1312, 1313 and 1314 of the processor integrated circuit in the second new product is converted into the low speed calculator 110 and the high speed calculator Let 120 be a program for dual DSP P14 to execute parallel arithmetic processing.
- dual DSP program P 14 is a program modified from low-speed DSP program P 10 as described in FIG. 2, and the processing performed by low-speed DSP program P 10 is Suppose that the program is divided into two programs, “compression processing” and “recording judgment processing”.
- the third connection step as shown in FIG.
- connection between the program memories 131 1 and 1312 and the low speed computing unit 110 and the connection between the program memories 1313 and 1314 and the high speed computing unit 120 are calculated.
- the processor P 12 for high-speed DSP can be converted to the program P 14 for dual DSP
- the changing process can be performed simultaneously with the mass production process of the second new product, and the time required for program change can be hidden. Also, there is no need to develop L S I for the fourth new product 14.
- the second new product 12 to the fourth new product 14 have been described as an example, as indicated in claims 12.
- the first new product 11 to the fourth new product 14 may be developed immediately. That is, when developing the first new product to the fourth new product, the low-speed DSP program P stored in the program memory in the program change step which is the second program change step in the third embodiment. 10 is changed to program P 14 for dual DSP, and each operation is performed in the first connection step and the second connection step which are the third connection step and the fourth connection step in the third embodiment.
- Device and program memory and data memory should be connected to execute the dual DSP program P14.
- the processor P 10 for low-speed DSP can be used as the program P 1 4 for dual DSP. Can be performed at the same time as the mass production process of the first new product, and the time required for program change can be concealed. It goes without saying that the third new product 13 to the fourth new product 14 can be developed using the same program as the second new product 12. As a result, by using the same processor integrated circuit 500 for both the third new product 13 and the fourth new product 14, the processor P 12 for high-speed DSP can be converted to the program P 14 for dual DSP. The changing process can be performed simultaneously with the mass production process of the third new product 13. Can hide the time required for the change.
- the dual DSP program P14 used for the fourth new product is a program changed from the low speed DSP program P10
- the high speed DSP program P12 It may be a modified program.
- the processor integrated circuit described in the third embodiment of the present invention as described above does not require a dedicated LSI in the first, second, third and fourth product developments, and consumes less power with the same LSI. Or we can advance the development of high-performance products.
- the example of product development using the processor integrated circuit configured by the low speed and high speed operation unit group shown in FIG. 2 has been described.
- a processor integrated circuit composed of n (n is a natural number of 2 or more) arithmetic operation unit groups may be used.
- the fourth new product can be a deferred model that can record at 6 ⁇ speed.
- the present invention is not limited to this.
- product development is performed based on the first new product. Also good.
- the second new product 13 in the third embodiment 13 and the fourth new product 14 are increasing the recording speed by high performance, adding sound quality correction and sound field processing, It is also possible to improve the function or change the function, such as changing to playback. Industrial applicability
- the processor integrated circuit according to the present invention is a processor configured on one chip LSI and is useful for improving audio product performance while maintaining program compatibility. It can also be applied to video processing applications.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Executing Machine-Instructions (AREA)
- Semiconductor Integrated Circuits (AREA)
- Microcomputers (AREA)
- Multi Processors (AREA)
- Advance Control (AREA)
Abstract
Description
Claims
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP04771636A EP1662376A4 (en) | 2003-08-07 | 2004-08-06 | INTEGRATED PROCESSOR SWITCHING AND PRODUCT DEVELOPMENT PROCESS WITH INTEGRATED PROCESSOR SWITCHING |
| JP2005513026A JP3887005B2 (ja) | 2003-08-07 | 2004-08-06 | プロセッサ集積回路 |
| US10/567,373 US20060206689A1 (en) | 2003-08-07 | 2004-08-06 | Processor integrated circuit and product development method using the processor integrated circuit |
| US12/588,673 US20100049944A1 (en) | 2003-08-07 | 2009-10-23 | Processor integrated circuit and product development method using the processing integrated circuit |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2003289410 | 2003-08-07 | ||
| JP2003-289410 | 2003-08-07 |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US12/588,673 Division US20100049944A1 (en) | 2003-08-07 | 2009-10-23 | Processor integrated circuit and product development method using the processing integrated circuit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2005015386A1 true WO2005015386A1 (ja) | 2005-02-17 |
Family
ID=34131557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2004/011661 Ceased WO2005015386A1 (ja) | 2003-08-07 | 2004-08-06 | プロセッサ集積回路及びプロセッサ集積回路を用いた製品開発方法 |
Country Status (7)
| Country | Link |
|---|---|
| US (2) | US20060206689A1 (ja) |
| EP (2) | EP2065808B1 (ja) |
| JP (2) | JP3887005B2 (ja) |
| CN (2) | CN100390728C (ja) |
| AT (1) | ATE485562T1 (ja) |
| DE (1) | DE602004029729D1 (ja) |
| WO (1) | WO2005015386A1 (ja) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6212073B2 (ja) * | 2015-06-29 | 2017-10-11 | ファナック株式会社 | プログラムの内容に応じて格納先を自動選択する機能を備えた数値制御装置 |
| US11476947B2 (en) * | 2019-05-24 | 2022-10-18 | Google Llc | Low power coherent receiver for short-reach optical communication |
| CN113485189A (zh) * | 2021-07-09 | 2021-10-08 | 绍兴光大芯业微电子有限公司 | 低速单片机实现代码高速运行且数据掉电不丢失的系统、方法、装置、存储器及其存储介质 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0452832A (ja) * | 1990-06-15 | 1992-02-20 | Mitsubishi Electric Corp | 演算制御システム |
| JPH04363704A (ja) * | 1991-02-25 | 1992-12-16 | Matsushita Electric Works Ltd | プログラマブルコントローラの演算プロセッサ |
| JPH05324455A (ja) * | 1992-04-07 | 1993-12-07 | Nec Corp | マルチプロセッサとメモリとのバス結合方式 |
| JPH0628173A (ja) * | 1992-07-07 | 1994-02-04 | Minolta Camera Co Ltd | プログラムの置換方法及び装置 |
| JPH08180027A (ja) * | 1994-12-21 | 1996-07-12 | Fuji Facom Corp | 調停回路 |
| JP2001014214A (ja) * | 1999-06-28 | 2001-01-19 | Matsushita Electric Ind Co Ltd | メモリ共有方法、およびこの方法を使用したマルチプロセッサ設備 |
| JP2002215597A (ja) * | 2001-01-15 | 2002-08-02 | Mitsubishi Electric Corp | マルチプロセッサ装置 |
Family Cites Families (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FI70682C (fi) * | 1984-06-20 | 1987-08-05 | Kemira Oy | Traeskyddsmedel och dess anvaendning som ytbehandlingsmedel. |
| US4727491A (en) * | 1984-06-27 | 1988-02-23 | Compaq Computer Corporation | Personal computer having normal and high speed execution modes |
| US4991169A (en) * | 1988-08-02 | 1991-02-05 | International Business Machines Corporation | Real-time digital signal processing relative to multiple digital communication channels |
| CA2069711C (en) * | 1991-09-18 | 1999-11-30 | Donald Edward Carmon | Multi-media signal processor computer system |
| US5748945A (en) * | 1996-05-31 | 1998-05-05 | International Business Machiens Corporation | Method for slave DMA emulation on a computer system bus |
| US6513057B1 (en) * | 1996-10-28 | 2003-01-28 | Unisys Corporation | Heterogeneous symmetric multi-processing system |
| JP2000029696A (ja) | 1998-07-08 | 2000-01-28 | Sony Corp | プロセッサおよびパイプライン処理制御方法 |
| US6415345B1 (en) * | 1998-08-03 | 2002-07-02 | Ati Technologies | Bus mastering interface control system for transferring multistream data over a host bus |
| US6240521B1 (en) * | 1998-09-10 | 2001-05-29 | International Business Machines Corp. | Sleep mode transition between processors sharing an instruction set and an address space |
| US20020034971A1 (en) * | 1999-02-08 | 2002-03-21 | Chienchung Chang | Data allocation for multiple applications on a microprocessor or dsp |
| DE19939763A1 (de) * | 1999-08-21 | 2001-02-22 | Philips Corp Intellectual Pty | Multiprozessorsystem |
| US6738840B1 (en) * | 1999-08-31 | 2004-05-18 | Koninklijke Philips Electronics N.V. | Arrangement with a plurality of processors having an interface for a collective memory |
| JP2001092662A (ja) * | 1999-09-22 | 2001-04-06 | Toshiba Corp | プロセッサコア及びこれを用いたプロセッサ |
| US6501999B1 (en) * | 1999-12-22 | 2002-12-31 | Intel Corporation | Multi-processor mobile computer system having one processor integrated with a chipset |
| JP4640880B2 (ja) | 2000-07-14 | 2011-03-02 | 国立大学法人東京工業大学 | マイクロプロセッサシステム |
| US6795875B2 (en) * | 2000-07-31 | 2004-09-21 | Microsoft Corporation | Arbitrating and servicing polychronous data requests in direct memory access |
| JP4915631B2 (ja) * | 2000-09-06 | 2012-04-11 | エスティー‐エリクソン、ソシエテ、アノニム | プロセッサ間通信システム |
| US6892266B2 (en) * | 2000-11-15 | 2005-05-10 | Texas Instruments Incorporated | Multicore DSP device having coupled subsystem memory buses for global DMA access |
| US6986066B2 (en) * | 2001-01-05 | 2006-01-10 | International Business Machines Corporation | Computer system having low energy consumption |
| JP2003186567A (ja) * | 2001-12-19 | 2003-07-04 | Matsushita Electric Ind Co Ltd | マイクロプロセッサ |
| US7069189B2 (en) * | 2002-12-31 | 2006-06-27 | Intel Corporation | Method and apparatus for controlling multiple resources using thermal related parameters |
-
2004
- 2004-08-06 WO PCT/JP2004/011661 patent/WO2005015386A1/ja not_active Ceased
- 2004-08-06 AT AT09155014T patent/ATE485562T1/de not_active IP Right Cessation
- 2004-08-06 US US10/567,373 patent/US20060206689A1/en not_active Abandoned
- 2004-08-06 DE DE602004029729T patent/DE602004029729D1/de not_active Expired - Lifetime
- 2004-08-06 CN CNB2004800227081A patent/CN100390728C/zh not_active Expired - Fee Related
- 2004-08-06 EP EP09155014A patent/EP2065808B1/en not_active Expired - Lifetime
- 2004-08-06 EP EP04771636A patent/EP1662376A4/en not_active Withdrawn
- 2004-08-06 JP JP2005513026A patent/JP3887005B2/ja not_active Expired - Fee Related
- 2004-08-06 CN CNB2007101851097A patent/CN100552655C/zh not_active Expired - Fee Related
-
2006
- 2006-10-20 JP JP2006286678A patent/JP2007073067A/ja active Pending
-
2009
- 2009-10-23 US US12/588,673 patent/US20100049944A1/en not_active Abandoned
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0452832A (ja) * | 1990-06-15 | 1992-02-20 | Mitsubishi Electric Corp | 演算制御システム |
| JPH04363704A (ja) * | 1991-02-25 | 1992-12-16 | Matsushita Electric Works Ltd | プログラマブルコントローラの演算プロセッサ |
| JPH05324455A (ja) * | 1992-04-07 | 1993-12-07 | Nec Corp | マルチプロセッサとメモリとのバス結合方式 |
| JPH0628173A (ja) * | 1992-07-07 | 1994-02-04 | Minolta Camera Co Ltd | プログラムの置換方法及び装置 |
| JPH08180027A (ja) * | 1994-12-21 | 1996-07-12 | Fuji Facom Corp | 調停回路 |
| JP2001014214A (ja) * | 1999-06-28 | 2001-01-19 | Matsushita Electric Ind Co Ltd | メモリ共有方法、およびこの方法を使用したマルチプロセッサ設備 |
| JP2002215597A (ja) * | 2001-01-15 | 2002-08-02 | Mitsubishi Electric Corp | マルチプロセッサ装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| EP2065808B1 (en) | 2010-10-20 |
| EP2065808A1 (en) | 2009-06-03 |
| CN100552655C (zh) | 2009-10-21 |
| JPWO2005015386A1 (ja) | 2006-10-05 |
| US20100049944A1 (en) | 2010-02-25 |
| JP2007073067A (ja) | 2007-03-22 |
| EP1662376A4 (en) | 2009-02-18 |
| JP3887005B2 (ja) | 2007-02-28 |
| ATE485562T1 (de) | 2010-11-15 |
| DE602004029729D1 (de) | 2010-12-02 |
| EP1662376A1 (en) | 2006-05-31 |
| CN101149716A (zh) | 2008-03-26 |
| CN1833221A (zh) | 2006-09-13 |
| CN100390728C (zh) | 2008-05-28 |
| US20060206689A1 (en) | 2006-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Renaudin et al. | ASPRO-216: a standard-cell QDI 16-bit RISC asynchronous microprocessor | |
| US5459843A (en) | RISC-type pipeline processor having N slower execution units operating in parallel interleaved and phase offset manner with a faster fetch unit and a faster decoder | |
| JPH09212358A (ja) | データ処理装置及びマイクロプロセッサ | |
| JP2003005958A (ja) | データ処理装置およびその制御方法 | |
| JPH05100948A (ja) | 2乗演算を実行する速度改良型データ処理システム及びその方法 | |
| TW201007465A (en) | A sharable memory architecture of cache in a embedded controller and a method of sharing memory | |
| WO2005015386A1 (ja) | プロセッサ集積回路及びプロセッサ集積回路を用いた製品開発方法 | |
| MX2009000728A (es) | Mecanismo de almacenamiento eficaz de direcciones de retorno de interrupcion. | |
| US20040064748A1 (en) | Methods and apparatus for clock domain conversion in digital processing systems | |
| JP2002032218A (ja) | マイクロプロセッサシステム | |
| US6931506B2 (en) | Electronic device for data processing, such as an audio processor for an audio/video decoder | |
| US10061737B2 (en) | Signal processing device, method of signal processing, storage medium, and electronic musical instrument | |
| JPS5824954A (ja) | アドレス制御方式 | |
| CN114296804B (zh) | 通过双模指令集架构可避免load_use冒险停顿的阵列处理器 | |
| JP2004362215A (ja) | プロセッサ及び半導体集積回路 | |
| JP2918019B2 (ja) | シングルチップマイクロプロセッサのテスト回路 | |
| JP2001142696A (ja) | マイクロプロセッサにおける命令供給装置および命令供給方法 | |
| JP3996982B2 (ja) | プロセッサ | |
| JP3531208B2 (ja) | ディジタル信号処理装置 | |
| JPH1091439A (ja) | プロセッサ | |
| JP3435252B2 (ja) | 情報処理装置 | |
| JPH10283186A (ja) | プロセッサ | |
| JPH09120284A (ja) | ディジタル信号処理装置 | |
| JP2001267909A (ja) | 信号処理装置及びその方法 | |
| JPH11328011A (ja) | 記憶制御装置および情報処理機器 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WWE | Wipo information: entry into national phase |
Ref document number: 200480022708.1 Country of ref document: CN |
|
| AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
| AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| WWE | Wipo information: entry into national phase |
Ref document number: 2005513026 Country of ref document: JP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 10567373 Country of ref document: US |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2004771636 Country of ref document: EP |
|
| WWP | Wipo information: published in national office |
Ref document number: 2004771636 Country of ref document: EP |
|
| WWP | Wipo information: published in national office |
Ref document number: 10567373 Country of ref document: US |