WO2014045500A1 - Lsi及びlsi製造方法 - Google Patents
Lsi及びlsi製造方法 Download PDFInfo
- Publication number
- WO2014045500A1 WO2014045500A1 PCT/JP2013/004155 JP2013004155W WO2014045500A1 WO 2014045500 A1 WO2014045500 A1 WO 2014045500A1 JP 2013004155 W JP2013004155 W JP 2013004155W WO 2014045500 A1 WO2014045500 A1 WO 2014045500A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- address
- core
- register
- lsi
- signal
- Prior art date
Links
- 238000004519 manufacturing process Methods 0.000 title claims description 15
- 238000000034 method Methods 0.000 claims description 34
- 238000012545 processing Methods 0.000 claims description 32
- 238000010586 diagram Methods 0.000 description 20
- 230000006870 function Effects 0.000 description 3
- 230000005236 sound signal Effects 0.000 description 3
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 2
- 229910052710 silicon Inorganic materials 0.000 description 2
- 239000010703 silicon Substances 0.000 description 2
- 101150046174 NIP2-1 gene Proteins 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
-
- 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
- G06F9/30145—Instruction analysis, e.g. decoding, instruction word fields
-
- 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
- G06F9/30098—Register arrangements
-
- 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
- G06F9/34—Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes
-
- 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
- G06F9/3889—Concurrent instruction execution, e.g. pipeline or look ahead using a plurality of independent parallel functional units controlled by multiple instructions, e.g. MIMD, decoupled access or execute
- G06F9/3891—Concurrent instruction execution, e.g. pipeline or look ahead using a plurality of independent parallel functional units controlled by multiple instructions, e.g. MIMD, decoupled access or execute organised in groups of units sharing resources, e.g. clusters
Definitions
- the present invention relates to an LSI having a plurality of IP cores and an LSI manufacturing method thereof.
- IP cores Intelligent Property Core
- the control of the IP core is realized by the CPU accessing the IP core control register arranged in the IP core.
- the same value is often written to a control register that controls the operation of each IP core.
- the CPU has to access the control registers for writing the same value one by one, which increases the load on the CPU.
- an LSI that can reduce the load on the CPU by operating two IP cores in parallel by accessing a common address that associates the two IP cores (for example, Patent Document 1).
- the LSI disclosed in Patent Document 1 includes two IP cores and one address decoder. Further, the IP core includes a plurality of control registers therein. Further, the same common address is assigned to the control register for writing the same value between the IP cores as an address for identifying the control register. When accessing each register, the CPU outputs an upper address (selection signal) and a lower address (common address). A plurality of IP cores can be specified by an upper address, and a plurality of control registers for writing the same value can be specified by a lower address.
- the present invention has been made in view of the above problems, and has an object to enable simultaneous access to a plurality of IP cores while reducing the CPU program development load.
- the LSI includes a plurality of registers, a plurality of IP cores that process input data, an address decoder that selects and activates the plurality of registers, and the IP used for processing the input data
- a CPU that outputs a system address signal designating a core register to the address decoder and writes the information of the input data to a register activated by the address decoder, the address decoder for processing the input data
- a register designated by the system address signal and another register in which the same information as the designated register is written are selected and made active.
- the LSI manufacturing method of the present invention includes a plurality of registers, a plurality of IP cores that process input data, an address decoder that selects and activates the plurality of registers, and the input data that is used for the processing
- a manufacturing method of an LSI comprising: a CPU that outputs a system address signal designating an IP core register to the address decoder and writes information of the input data to a register activated by the address decoder; Generating a system address map in which a system address for the CPU to identify the plurality of registers and an IP address for the address decoder to identify the plurality of registers are assigned, and one system address Multiple registers activated by signal Generating register grouping information that constitutes a group including the address, and using the system address map and the register grouping information, address decoding that associates a plurality of system addresses included in the same group with the IP address Generating information, and manufacturing the address decoder based on the address decoding information.
- the LSI of the present invention includes an address decoder that selects and activates a control register designated by one system address signal and another register in which the same information as the designated control register is written. Even when the number of IP cores to be used increases, a plurality of control registers can be accessed with a single system address signal without increasing the CPU program development load.
- FIG. 1 is a configuration diagram of an LSI according to a first embodiment.
- FIG. 3 is a diagram showing an operation flowchart of the LSI according to the first embodiment.
- FIG. 3 is a diagram showing a system address map according to the first embodiment.
- FIG. 5 is a diagram showing address decode information of each IP core according to the first embodiment.
- FIG. 5 is a configuration diagram of an LSI according to a third embodiment.
- FIG. 6 is a configuration diagram of an LSI according to a fourth embodiment.
- FIG. 10 is a diagram showing an address decoding information generation flowchart according to the fifth embodiment.
- FIG. 20 shows a list of instance names according to the fifth embodiment.
- FIG. 10 shows an address map of each IP core according to the fifth embodiment.
- FIG. 10 shows control register grouping information according to the fifth embodiment.
- FIG. 10 is a flowchart illustrating an operation for assigning an IP address and a system address according to the fifth embodiment.
- FIG. 1 is a configuration diagram of an LSI according to the first embodiment.
- FIG. 2 is a flowchart showing the operation of the LSI according to the first embodiment.
- FIG. 3 is a diagram showing a system address map according to the first embodiment.
- FIG. 4 is a diagram showing address decode information of each IP core according to the first embodiment.
- FIG. 5 is a diagram illustrating an example (when path 1 is selected) in which the IP cores connected to each other according to the first embodiment process input video signal data.
- the LSI 1 includes a CPU 2, an operation mode control circuit 6, an address decoder 3, an IP core (4ipa1, 4ipa2, 4ipb, 4ipc) (hereinafter, each IP core is collectively referred to as “IP Core 4 ”) and selector 5.
- IP Core 4 an IP core (4ipa1, 4ipa2, 4ipb, 4ipc) (hereinafter, each IP core is collectively referred to as “IP Core 4 ”) and selector 5.
- IP Core 4 IP core 4ipa1, 4ipa2, 4ipb, 4ipc)
- selector 5 selector 5
- the LSI 1 adjusts the image quality of the input video signal data 105 (input data) output from the video signal input device 7 based on various information such as format and resolution input via a user interface (not shown).
- the output video signal data 106 or 107 is output.
- the CPU 2 accesses an IP core 4 to be described later and causes the IP core 4 to execute processing of the input video signal data 105. Further, the CPU 2 outputs a system address signal 100 and a write enable signal 102 when accessing.
- the system address signal 100 is a signal indicating a system address.
- the system address is a value that identifies the location of all elements (including a memory (not shown)) viewed from the CPU 2.
- the write enable signal 102 is a signal indicating whether an access to a control register by an address decoder 3 described later is a read or a write.
- the write enable signal becomes Read when the access to the control register is read, and becomes Write when the access is performed.
- the control register accessed by the address decoder 3 is in a readable or writable state. This state is called active.
- the operation mode control circuit 6 outputs an operation mode signal 101.
- the operation mode signal 101 is a signal that determines a combination of IP cores 4 to be used when the input video signal data 105 is processed by the LSI 1.
- the address decoder 3 receives the system address signal 100, the operation mode signal 101, and the write enable signal 102, and outputs an IP address signal 103 and a chip select signal 104.
- the IP address signal 103 is a signal indicating an IP address.
- the IP address is an address for the address decoder 3 to identify the location of a control register of the IP core 4 described later.
- the address decoder 3 allows the CPU 2 to access a desired IP core 4.
- the IP core 4 is a functional block that executes a specific process, and is, for example, an image processing circuit or an audio processing circuit.
- the IP core 4 has a control register therein, and the CPU 2 processes the input video signal data 105 by writing information on the format of the input video signal data 105 into the control register.
- the IP core 4 receives the IP address signal 103, the chip select signal 104, and the write enable signal 102 and activates the control register.
- the CPU 2 executes a write process on the activated control register. For example, when converting the input video signal data 105 into a desired format, the CPU 2 writes the desired format in the activated control register.
- the selector 5 switches the connection relationship between the plurality of IP cores 4.
- the selector 5 receives the operation mode signal 101 and changes the processing path of the input video signal data 105.
- Step 201 is an operation mode switching step.
- the operation mode switching step is a step in which the selector 5 switches the combination of the IP cores 4 that process the input video signal data 105.
- step 201 the operation mode control circuit 6 outputs an operation mode signal 101 to the selector 5 and the address decoder 3.
- the selector 5 selects and connects the designated IP core 4 from the plurality of IP cores 4 to be used based on the operation mode signal 101. For example, when the operation mode of the operation mode signal 101 is the path 1, the selector 5 connects the IP core 4ipa1 and the IP core 4ipb, and further connects the IP core 4ipa2 and the IP core 4ipc. The selector 5 connects the IP core 4ipa1 and the IP core 4ipc and connects the IP core 4ipa2 and the IP core 4ipb when the operation mode is the path 2.
- the combination of IP cores 4 to be connected corresponds to the combination of IP cores 4 having a control register that is activated by one system address signal described in step 202 described later.
- the selector 5 connects the IP core 4ipa1 and the IP core 4ipb.
- the process proceeds to step 202.
- Step 202 is a step for setting a register.
- the register setting is an operation of writing information related to the input video signal data 105 to the control register activated by the CPU 2.
- step 202 the CPU 2 outputs the system address signal 100 and the write enable signal 102 to the address decoder 3. Further, the CPU 2 outputs a write enable signal 102 to the IP core 4.
- the address decoder 3 receives the system address signal 100 and the write enable signal 102 from the CPU 2 and the operation mode signal 101 from the operation mode control circuit 6, and outputs the IP address signal 103 and the chip select signal 104 to the IP core 4. To do.
- the address decoder 3 receives the system address signal 100, the write enable signal 102 and the operation mode signal 101 and outputs the IP address signal 103 and the chip select signal 104 will be described in detail with reference to FIGS. To do.
- FIG. 3 is a diagram showing a system address map according to the first embodiment.
- the address decoder 3 determines the IP address signal 103 and the chip select signal 104 using the system address map 301.
- the instance name of the IP core 4 is associated with the system address of each control register.
- the instance name is a name assigned to identify the IP core 4.
- the instance names of the IP cores 4ipa1, 4ipa2, 4ipb, and 4pc correspond to ipa1, ipa2, ipb, and ipc, respectively.
- the control register name represents the name of the control register in the IP core 4 and is given a name based on the value to be written. For example, a control register name “format” is assigned to the control register in which the format of the input video signal 105 is written.
- a control register name “resolution” is assigned to the control register to which the resolution is written
- a frame register is assigned to the control register to which the frame rate is written.
- the IP address is a value for identifying the location of the control register by the address decoder 3 and is assigned to the control register in each IP core 4. This IP address is different from the system address as long as the control register can be identified within the IP core, and the same value may be assigned to the control register between different IP cores. For example, in FIG. 3, IP addresses 0 to 2 are assigned to the control registers in each IP core 4. Based on the system address map 301, the address decode information 302 shown in FIG. 4 is created.
- FIG. 4 is a diagram showing the address decode information 302.
- a plurality of system addresses are associated with the IP addresses of the IP cores 4 respectively.
- a combination (group) of a plurality of system addresses corresponding to one IP address is a combination of control registers that are accessed by one system address signal and write the same value.
- operation mode signal path 1
- the system address “2, 7” is stored in the control register frameate in the IP core 4ipa1 (instance name: ipa1) and in the IP core 4ipb (instance name: ipb). ) Control register framerate. These control registers indicate that the same value is written by the CPU 2. Similarly, other system addresses “0, 6” and “1” indicate combinations of registers to which the same value is written.
- the address decode information 302 has a combination of a plurality of system addresses according to the operation mode, even when the IP core to be used is switched, the program of the operation mode control circuit 6 is not rewritten. It is possible to change the combination of control registers to be accessed simply by changing the value of the operation mode signal 101.
- the address decoder 3 refers to the column “When Write” of the address decode information 302 when the input write enable signal 102 indicates Write, and reads “Read” when the input write enable signal 102 indicates Read. Refer to the “hour” column.
- the address decoder 3 outputs “2” as the corresponding IP address signal 103 to the IP core 4ipa1 and outputs “1” as the IP address signal 103 to the IP core 4ipb. Further, when the operation mode is path 1, the system addresses “2, 7” are in the same set, so even if the system address signal 102 is 7, the same IP address has the same IP core. (4ipa1, 4ipb). As described above, by setting the system address as a set and address decoding information 302, the address decoder 3 changes from the system address signal 100 indicating one system address to the IP address signal 103 for designating a plurality of control registers. It becomes possible to convert.
- the address decoder 3 can be realized using hardware such as an electronic circuit based on the correspondence relationship of the decode information 302.
- the address decoder 3 converts the system address signal 100, the write enable signal 102, and the operation mode signal 101 and outputs the IP address signal 103 to the corresponding IP core 4.
- the address decoder 3 outputs an enable signal as the chip select signal 104 to make it active with respect to the IP core 4 having the control register designated by the IP address signal 103.
- the IP core 4 receives the IP address signal 103 and the chip select signal 104 and activates the control register.
- the IP core 4 Upon receiving the IP address signal 103 and the chip select signal 104 output from the address decoder 3, the IP core 4 activates the control register designated by the IP address signal 103.
- step 202 for setting the writing to the control registers ends, and the process proceeds to step 203.
- Step 203 is a step in which the IP core 4 processes the input video signal data 105.
- the input video signal data 105 is input to the IP core 4 from the video signal input device 7 outside the LSI 1.
- the input video signal data 105 is processed by the IP core 4, sequentially processed by the plurality of IP cores 4 connected in step 201, and finally output as output video signal data 106 or 107.
- the IP core 4 ipa 1 and the IP core 4 ipb are connected by the selector 5
- the input video signal data 105 sequentially passes through the IP cores 4 ipa 1 and 4 ipb and is output as output video signal data 106.
- the IP core 4ipa2 and the IP core 4ipc are connected, the IP core 4ipa2 and 4ipc sequentially pass through and output video signal data 107 is output. At this time, a series of operations of the LSI 1 is completed.
- the operation mode control circuit 6 outputs the operation mode signal 101 indicating the path 1 to the selector 5 and the address decoder 3.
- the selector 5 selects and connects a combination of IP cores 4 set in advance based on the value of the operation mode signal 101.
- the selector 5 connects the IP core 4ipa1 and the IP core 4ipb, and connects the IP core 4ipa2 and the IP core 4ipc.
- the operation mode control circuit 6 also outputs an operation mode signal 101 indicating the path 1 to the address decoder 3.
- the CPU 2 that has received the register write command for processing the input video signal data 105 sends the address decoder 3 the system address signal 100 indicating 0 and the write enable signal 102 indicating Write. Is output.
- the IP addresses in the row of the system address 0 are the IP address 0 in the 4ipa1 table and the IP address “0” in the 4ipb table.
- the IP address “0” of the IP core 4ipa1 corresponds to the control register format
- the IP address “0” of the IP core 4ipb corresponds to the control register format. Therefore, the address decoder 3 converts the system address signal 100 indicating 0 into the IP address signal 103 indicating 0 and outputs the IP address signal 103 to the IP core 4ipa1 and the IP core 4ipb, and designates the control register format to be activated. Further, the address decoder 3 outputs enable as the chip select signal 104 to the control register format of the designated IP core 4ipa1 and IP core 4ipb, and activates the control register format.
- the IP core 4ipa1 and the IP core 4ipb having the control register format set to active receive the write enable signal 102 indicating Write and determine that writing is performed.
- the CPU 2 receives information (in this case, input video signal data 105) via a signal line (not shown) directly connected from the CPU 2 to the control register format of the IP core 4ipa1 and the IP core 4ipb.
- the information of the signal data 105 is A.
- A indicates the format of the input video signal data 105 such as MPEG.
- the CPU 2 outputs 1 and 2 as the system address signal 100 and writes values B and C indicating the information of the input video signal data 105 to the remaining control register resolution and control register framerate, respectively.
- the address decoder 3 outputs the chip select signal 104 to the control register resolution of the designated IP core 4ipa1 to make it active. Subsequently, the value B is written by the CPU 2 in the control register format of the IP core 4ipa1 that is activated. Further, when a signal indicating 2 is input to the address decoder 3 as the system address signal 100, 2 is output as the IP address signal 103 to the IP core 4ipa1, and 1 is output as the IP address signal 103 to the IP core 4ipb. Specifies the control register framerate. The address decoder 3 outputs enable as a chip select signal 104 to the designated control register framerate of the IP core 4ipa1 and the IP core 4ipab to make it active.
- a value C is written by the CPU 2 to the activated control register framerate.
- B is a value representing the resolution
- C is a value representing the frame rate.
- the value written to the control register that is, the value A indicating the format, the value B indicating the resolution, and the frame rate.
- the value C indicating the value changes depending on the type of the input video signal data 105.
- the input video signal data 105 sequentially passes through the IP core 4ipa1 and the IP core 4ipb having the control register in which the same value is written to adjust the image quality.
- the IP core 4ipa1 is an IP core 4 having a function of suppressing noise
- the IP core 4ipb is an IP core 4 that converts color tone
- the input video signal data 105 is an output video signal in which noise is suppressed and color tone is converted. Output as data 106.
- the LSI 1 has been described as executing the processing of the input video signal data 105 using the IP core 4ipa1 and the IP core 4ipb.
- the LSI1 uses the IP core 4ipa2 and the IP core 4ipc.
- the processing of the input signal data 105 can be executed.
- the CPU 2 outputs a system address signal 100 indicating 3, 4, and 5 to the address decoder 3 as in the example of writing to the IP core 4ipa1 and the IP core 4ipb.
- the address decoder 3 that has received the system address signal 100 refers to the 4ipa2 and 4ipc tables of the address decode information 302 and activates the control registers format, resolution, and frameate of the IP core 4ipa2 and the IP core 4ipc, respectively. .
- the CPU 2 executes writing to the activated control register.
- the input video signal data 105 sequentially passes through the IP core 4ipa2 and the IP core 4ipc and is output as output signal data 107.
- the LSI 1 can also execute processing of the input signal data 105 using the IP core 4ipa1, the IP core 4ipc, the IP core 4ipa2, and the IP core 4ipb (not shown).
- the operation mode control circuit 6 outputs an operation mode signal indicating the path 2 to the selector 5.
- the selector 5 connects the IP core 4ipa1 and the IP core 4ipc, and the IP core 4ipa2 and the IP core 4ipb.
- the address decoder 3 refers to the path 2 column of the address decode information 302.
- the IP core 4 has been described as executing video signal processing. However, the present invention is not limited to this, and any device that executes data processing may be used. For example, the IP core 4 may execute audio signal processing. At that time, the LSI 1 processes the audio signal data, not the input video signal data 105.
- the four IP cores 4 are used.
- a plurality of control registers may be selected by one system address signal.
- the number of IP cores may be two or more. Note that when the LSI 1 according to the first embodiment switches the IP core 4 used in response to the operation mode signal, at least three IP cores 4 are required.
- the address decoder 3 receives one system address signal and writes one or two control registers to write the same value.
- the present invention is not limited to this. You may comprise so that it may write in three or more control registers.
- the address decode information 302 three or more system addresses are associated with one IP address row.
- the LSI 1 since the LSI 1 according to the first embodiment sets the combination of the IP core 4 and the control register that are accessed by one system address signal as the operation mode signal 101 in advance, the CPU 2 stores the control register. There is no need to prepare selection signals as many as the number of combinations, and the burden of CPU 2 program development can be reduced.
- the address decoder 3 operates based on address decode information in which a plurality of system addresses are associated with a single IP address. Can be accessed.
- Embodiment 2 The LSI 1 according to the second embodiment will be described with reference to FIGS.
- FIG. 6 is a diagram showing an example (when path 1 is selected) in which IP cores connected to each other according to the second embodiment process input video signal data in parallel.
- FIG. 7 is a diagram showing an example (when path 2 is selected) in which IP cores connected to each other according to the second embodiment process input video signal data in parallel. Note that portions corresponding to the configuration of the LSI 1 of the second embodiment are denoted by the same reference numerals as those in FIGS. 1 and 5 and description thereof is omitted.
- the values A, B, C, D, E, and F in FIGS. 6 and 7 indicate values to be written in the control registers such as the format and resolution of the input video signal data 108 and 109. It is shown that the same value is set for registers marked with.
- the LSI 1 according to the second embodiment processes the input video signal data 108 and the input video signal data 109 in parallel.
- an operation in which the LSI 1 processes the input video signal data 108 and 109 shown in FIGS. 6 and 7 in parallel will be described.
- FIG. 6 is an example when the operation mode signal 101 indicates the path 1.
- the operation mode control circuit 6 first outputs the operation mode signal 101 indicating the path 1.
- the selector 5 Upon receiving the operation mode signal 101 indicating the path 1, the selector 5 connects the IP core 4ipa1 and the IP core 4ipb. Furthermore, the selector 5 connects the IP core 4ipa2 and the IP core 4ipc. Since the write operation is the same as that of the first embodiment, the description is omitted.
- the input video signal data 108 and the input video signal data 109 are video signal data having different formats, resolutions, and frame rates, the CPU 2 needs to write values corresponding to the respective input video signals to the IP core 4. is there.
- the input video signal data 108 sequentially passes through the IP cores 4 ipa 1 and 4 ipb and is output as the input video signal data 110. Further, the input video signal data 109 sequentially passes through the IP cores 4 ipa 2 and 4 ipc and is output as output video signal data 111.
- FIG. 7 shows an example when the operation mode signal 101 indicates the path 2. Since the connection relationship between the front and rear IP cores 4 is switched by the selector 5, the combination of the IP cores 4 is different from that of the route 1.
- the selector 5 that has received the operation mode signal 101 indicating the path 2 connects the IP core 4ipa1 and the IP core 4ipc. Further, the IP core 4ipa2 and the IP core 4ipb are connected.
- the input video signal data 108 sequentially passes through the IP cores 4ipa1 and 4ipc and is output as output video signal data 110. Further, the input video signal data 109 sequentially passes through the IP cores 4 ipa 2 and 4 ipb and is output as output video signal data 111.
- the LSI 1 controls access based on the operation mode signal 101 output from the operation mode control circuit 6 even when a plurality of input video signal data 108 and 109 are input.
- the combination of registers for use can be determined, and a plurality of input video signal data can be processed in parallel.
- FIG. 8 is a block diagram of an LSI according to the third embodiment. Note that portions corresponding to the configuration of the LSI 1 of the third embodiment are denoted by the same reference numerals as those in FIG. 1 and description thereof is omitted.
- each IP core 4 is provided with an address decoder 31, 32, 33, 34 for each IP core. Specifically, a 4ipa1 address decoder 31, an IP core 4ipa1, 4ipa2 address decoder 32, an IP core 4ipa2, 4ipb address decoder 33, an IP core 4ipb, 4ipc address decoder 34, and an IP core 4ipc are connected.
- the CPU 2 outputs a system address signal 100 and a write enable signal 102 to all the address decoders 31, 32, 33, 34. Further, the operation mode control circuit 6 outputs the operation mode signal 101 to all the address decoders 31, 32, 33, 34.
- each IP core address decoder 31, 32, 33, 34 receives the system address signal 100, the write enable signal 102, and the operation mode signal 101, and is connected to the IP core 4 connected thereto.
- each of the IP address decoders 31, 32, 33 and 34 receives the system address signal 100 and refers to the address decode information 302 corresponding to the connected IP core 4.
- the address decoders 31, 32, 33, and 34 output the IP addresses obtained by referring to the address decode information 302 to the IP cores 4 as IP address signals (103a1, 103a2, 103b, and 103c).
- the control register is designated and a chip select signal (104a1, 104a2, 104b, 104c) is output to activate the selected control register.
- the CPU 2 writes a value indicating information of the input video signal data 105 to the activated control register. After the writing to all the control registers is completed, the LSI 1 starts processing the input video signal data 105.
- the LSI 1 uses the address decode information 302 in the operation mode even when the address decoders 31, 32, 33, and 34 corresponding to the respective IP cores 4 are provided. It is possible to execute register writing according to the above.
- FIG. 9 is a block diagram of an LSI according to the fourth embodiment. Note that portions corresponding to the configuration of the LSI 1 of the fourth embodiment are denoted by the same reference numerals as those in FIG. 1 and description thereof is omitted.
- the LSI 1 does not include the operation mode control circuit 6, and the operation mode signal 101 is output from the CPU 2.
- the CPU 2 outputs an operation mode signal 101 to the address decoder 3 and the selector 5.
- the selector 5 determines the combination of the IP cores 4 based on the operation mode signal 101, and connects the IP cores 4 of the determined combination.
- the address decoder 3 receives the operation mode signal 101 and outputs the IP address signal 103 and the chip select signal 104 to the IP core 4 according to the address decode information 302. Note that the operation of converting the operation mode signal 101 into the IP address signal 103 and the chip select signal 104 by the address decoder 3 is the same as the operation of the LSI 1 according to the first embodiment, and is therefore omitted.
- the LSI 1 since the LSI 1 according to the fourth embodiment is configured to output the operation mode signal 101 from the CPU 2, a plurality of control registers in each IP core 4 are provided without separately providing the operation mode control circuit 6. Can be accessed.
- FIG. 10 is a diagram showing an address decoding information generation flowchart according to the fifth embodiment.
- FIG. 11 is a diagram showing a list of instance names according to the fifth embodiment.
- FIG. 12 is an address map of each IP core according to the fifth embodiment.
- FIG. 13 shows register grouping information for control according to the fifth embodiment.
- FIG. 14 is a flowchart for explaining an operation of assigning an IP address and a system address according to the fifth embodiment.
- the LSI 1 according to the first embodiment generates an RTL description for each element such as the address decoder 3 and the IP core 4 according to the first embodiment, and combines them into a gate description level from a logic circuit in a step called logic synthesis. After being converted into a netlist, it is converted into a physical layout structure. The LSI 1 is created by injecting and baking the layout pattern thus created on a silicon wafer.
- a method for generating the address decode information 302 that is a characteristic part of the LSI 1 according to the first embodiment will be described in detail.
- RTL represents a logic circuit using a hardware description language such as HDL (Hardware DescriptionwLanguage), which is a kind of computer language for designing an integrated circuit. Is expressed at a level combined with a register.
- the address decode information 302 is generally automatically generated by a semiconductor design support apparatus or the like. Furthermore, in the description of the method of manufacturing the LSI 1 of the fifth embodiment, the same reference numerals are given to the portions corresponding to the configurations described in FIGS. 1 and 5 to 8, and the description thereof is omitted.
- FIG. 10 is a diagram showing an address decoding information generation flowchart.
- the address decode information 302 first, in step 030, the system address map 301 is generated.
- the IP address and system address are assigned using the system address map 301 created at step 030, and address decode information 302 is generated.
- the system address map 301 associates an instance name of the IP core 4 to be used, a control register name, and an IP address with the system address.
- FIG. 11 shows an instance name list 303 of the IP core 4.
- the instance name list 303 of the IP core 4 is a correspondence table between IP names and instance names.
- the IP name is a name assigned to the IP core 4, and the same name is given to the IP core 4 having the same function.
- the instance name is a name assigned to identify the IP core 4 to be used.
- FIG. 12 is an address map of each IP core 4.
- the address map of the IP core 4 associates the control register name of the IP core 4 with the IP address.
- the instance name list 303 of the IP core 4 and the address map 304 of each IP core 4 are manually created as a text file or the like.
- the system address map 301 is generated based on the instance name list 303 prepared as described above and the address map 304 of each IP core 4.
- step 060 will be described.
- FIG. 13 shows an example of the control register grouping information 305.
- the control register grouping information 305 indicates a group of control registers in which the CPU 2 writes the same value. Further, the control register grouping information 305 is classified for each operation mode. The combination of control registers in each group differs depending on the operation mode. The LSI designer determines this group based on the connection configuration and processing contents of the IP core 4 based on the combination of control registers. For example, the upper table of FIG. 13 shows control register grouping information created assuming that the operation mode is path 1, that is, IP core 4ipa1 and IP core 4ipb and IP core 4ipa2 and IP core 4ipc are connected. 305.
- the address decoder 3 uses a single system address signal to control a plurality of controls that each of the IP cores 4ipa1 and 4ipb has.
- the register format can be accessed. Further, in the control register grouping information 305, when the operation mode is path 2, it indicates that the selector 5 connects the IP core 4ipa1 and the IP core 4ipc and the IP core 4ipa2 and the IP core 4ipb.
- the instance name before “.” Means the instance name, and the part after “.” Means the control register name.
- the control register grouping information 305 is manually created as a text file or the like.
- the address decode information 302 is generated using the system address map 301 generated in step 030 and the control register grouping information 305.
- FIG. 14 is a flowchart for explaining an operation of assigning an IP address and a system address.
- step 061 one IP address is selected with reference to the system address map 301.
- the IP address obtained here is A.
- Step 062 the system address “when reading” is assigned to the IP address A in accordance with the system address map 301.
- step 063 one operation mode is selected.
- step 064 the control register grouping information 305 of the operation mode selected in step 063 is referenced to search the system address map 301 for a group to which the system address corresponding to the IP address A belongs.
- G is a group obtained by searching.
- step 065 system addresses of all control registers belonging to the group G are assigned to the IP address A, and address decoding information for “Write” is created.
- step 066 if the IP address A and the system address of the group to which the IP address A belongs are associated with all the operation modes, the process proceeds to step 067. If the IP address A is not associated with the system address of the group to which the IP address A belongs for all the operation modes, the process returns to step 063. When returning to step 063, an operation mode not yet assigned to IP address A is selected, and the operations from step 063 to step 066 are repeated.
- step 067 when the correspondence between the system addresses of “Read time” and “Write time” is completed for all IP addresses, the process is terminated. On the other hand, if the system address association of “Read” and “Write” is not completed, the process returns to the process 061. When returning to Step 061, Steps 061 to 066 are repeated, and system addresses are associated with IP addresses that have not yet been associated with system addresses.
- the address decode information 302 is generated by associating system addresses with all IP addresses.
- the address decode information 302 generated in this way determines the necessary input / output port width, generates an RTL description of the IP address signal and the chip select signal 104, and becomes the RTL description of the address decoder 3.
- This RTL description is logically synthesized into a net list, converted into a physical layout structure, and finally burned onto a silicon wafer to become LSI 1.
- the IP core 4 to be controlled is controlled. Even if this increases, the program development burden on the CPU 2 can be reduced.
- the LSI 1 manufacturing method according to the fifth embodiment has been described by taking the LSI 1 manufacturing method for processing the input video signal data 105 as an example.
- the LSI 1 manufacturing method for processing audio signal data or the like may be used, and the present invention is not limited thereto.
- the manufacturing method of the LSI 1 that processes the input video signal data 105 is not limited.
- the LSI manufacturing method according to the first embodiment has been described.
- the LSI 1 according to the second to fourth embodiments can be manufactured in the same procedure.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Microcomputers (AREA)
- Advance Control (AREA)
- Executing Machine-Instructions (AREA)
- Semiconductor Integrated Circuits (AREA)
- Multi Processors (AREA)
- Small-Scale Networks (AREA)
Abstract
Description
以下、図1から図5を用いて本発明の実施の形態1に係るLSIについて、映像信号データを処理するLSIを例に説明する。図1は実施の形態1に係るLSIの構成図である。図2は実施の形態1に係るLSIの動作フローチャートを示す図である。図3は実施の形態1に係るシステムアドレスマップを示す図である。図4は実施の形態1に係る各IPコアのアドレスデコード情報を示す図である。図5は実施の形態1に係る互いに接続されたIPコアが入力映像信号データを処理する例(経路1選択時)を示す図である。
図6、図7を用いて実施の形態2に係るLSI1について説明する。
以下、図8を用いて実施の形態3に係るLSIについて説明する。
以下、図9を用いて実施の形態4に係るLSIについて説明する。図9は実施の形態4に係るLSIの構成図である。なお、実施の形態4のLSI1の構成に相当する部分には図1と同一符号を付してその説明を省略する。
以下、図10から図14を用いて実施の形態5に係るLSIの製造方法を説明する。図10は実施の形態5に係るアドレスデコード情報生成フローチャートを示す図である。図11は実施の形態5に係るインスタンス名の一覧を示す図である。図12は実施の形態5に係る各IPコアのアドレスマップである。図13は実施の形態5に係る制御用レジスタグループ化情報である。図14は実施の形態5に係るIP用アドレスとシステムアドレスを割り当てる動作を説明するフローチャートを示すである。
Claims (8)
- 複数のレジスタを有し、入力データを処理する複数のIPコアと、
前記複数のレジスタを選択してアクティブとするアドレスデコーダと、
前記入力データの処理に使用する前記IPコアのレジスタを指定するシステムアドレス信号を前記アドレスデコーダに出力し、前記アドレスデコーダによってアクティブとされたレジスタに前記入力データの情報を書き込むCPUと
を備え、
前記アドレスデコーダは、前記入力データの処理に使用する前記IPコアのレジスタのうち、前記システムアドレス信号により指定されたレジスタと、該指定されたレジスタと同じ情報が書き込まれる他のレジスタとを選択してアクティブとすることを特徴とするLSI。 - 前記入力データの処理に使用するIPコアの組み合わせを指定する動作モード信号を前記アドレスデコーダに出力する動作モード制御回路を有し、
前記アドレスデコーダは、前記動作モード信号に基づいて前記入力データの処理に使用する前記IPコアの組み合わせを決定し、決定した該使用するIPコアのレジスタのうち、前記システムアドレス信号により指定されたレジスタと、該指定されたレジスタと同じ情報が書き込まれる他のレジスタとを選択してアクティブとすることを特徴とする請求項1に記載のLSI。 - 前記CPUは、入力データの処理に使用するIPコアの組み合わせを指定する動作モード信号を前記アドレスデコーダに出力し、
前記アドレスデコーダは、前記動作モード信号に基づいて前記入力データの処理に使用する前記IPコアの組み合わせを決定し、決定した該使用するIPコアのレジスタのうち、前記システムアドレス信号により指定されたレジスタと、該指定されたレジスタと同じ情報が書き込まれる他のレジスタとを選択してアクティブとすることを特徴とする請求項1に記載のLSI。 - 前記アドレスデコーダは、
前記CPUが前記レジスタを識別するための複数のシステムアドレスを含むグループと、該アドレスデコーダが前記レジスタを識別するためのIP用アドレスと、
前記システムアドレスと前記IP用アドレスとを対応づけたアドレスデコード情報とに基づいて、
前記CPUから出力されたシステムアドレス信号が示すシステムアドレスに対応するIP用アドレスを特定し、特定した該IP用アドレスが示すレジスタをアクティブとすることを特徴とする請求項1から請求項3のいずれかに記載のLSI。 - 前記アドレスデコード情報は、
前記動作モード信号が示す動作モードごとに前記グループに含まれるシステムアドレスの組み合わせを異なるものとし、該グループと前記IP用アドレスとを対応づけたことを特徴とする請求項4に記載のLSI。 - 前記動作モード信号を受けて、前記入力データの処理に使用するIPコア同士を接続するセレクタを備えることを特徴とする請求項2から請求項5のいずれかに記載のLSI。
- 前記複数のIPコアは、複数の前記入力データを並列処理することを特徴とする請求項1から請求項6のいずれかに記載のLSI。
- 複数のレジスタを有し、入力データを処理する複数のIPコアと、
前記複数のレジスタを選択してアクティブとするアドレスデコーダと、
前記入力データの処理に使用する前記IPコアのレジスタを指定するシステムアドレス信号を前記アドレスデコーダに出力し、前記アドレスデコーダによってアクティブとされたレジスタに前記入力データの情報を書き込むCPUと
を備えるLSIの製造方法において、
各レジスタに対して、前記CPUが前記複数のレジスタを識別するためのシステムアドレスと、前記アドレスデコーダが前記複数のレジスタを識別するためのIP用アドレスとを割り当てたシステムアドレスマップを生成するステップと、
1つのシステムアドレス信号でアクティブとする複数のレジスタを含むグループを構成したレジスタグループ化情報を生成するステップと、
前記システムアドレスマップと前記レジスタグループ化情報とを用いて、同じグループに含まれる複数のシステムアドレスと前記IP用アドレスを対応付けたアドレスデコード情報を生成するステップと、
前記アドレスデコード情報に基づいて前記アドレスデコーダを生成するステップとを有することを特徴とするLSIの製造方法。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE112013004631.8T DE112013004631T5 (de) | 2012-09-21 | 2013-07-04 | LSl und LSl-Herstellungsverfahren |
US14/429,157 US9424040B2 (en) | 2012-09-21 | 2013-07-04 | LSI and LSI manufacturing method |
CN201380048530.7A CN104641364A (zh) | 2012-09-21 | 2013-07-04 | Lsi和lsi制造方法 |
JP2014536562A JP5835498B2 (ja) | 2012-09-21 | 2013-07-04 | Lsi及びlsi製造方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012-207726 | 2012-09-21 | ||
JP2012207726 | 2012-09-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2014045500A1 true WO2014045500A1 (ja) | 2014-03-27 |
Family
ID=50340843
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/004155 WO2014045500A1 (ja) | 2012-09-21 | 2013-07-04 | Lsi及びlsi製造方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US9424040B2 (ja) |
JP (1) | JP5835498B2 (ja) |
CN (1) | CN104641364A (ja) |
DE (1) | DE112013004631T5 (ja) |
WO (1) | WO2014045500A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104597832A (zh) * | 2014-12-31 | 2015-05-06 | 浙江中控研究院有限公司 | 一种基于amba总线的plc程序调度器ip核 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH02254496A (ja) * | 1989-03-29 | 1990-10-15 | Yamaha Corp | 楽音発生装置 |
JPH0764957A (ja) * | 1993-08-23 | 1995-03-10 | Mitsubishi Electric Corp | タイマ装置 |
JPH11353265A (ja) * | 1998-06-05 | 1999-12-24 | Mitsubishi Electric Corp | 情報伝達装置 |
JP2004362157A (ja) * | 2003-06-03 | 2004-12-24 | Toshiba Corp | 半導体装置、そのアドレス割り付け方法、および半導体装置の制御方法 |
JP2008097372A (ja) * | 2006-10-12 | 2008-04-24 | Matsushita Electric Ind Co Ltd | システム制御装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4530042B2 (ja) * | 2005-05-12 | 2010-08-25 | 日本電気株式会社 | プロセッサアレイ、プロセッサエレメント複合体、マイクロ命令制御装置およびマイクロ命令制御方法 |
WO2006129722A1 (ja) * | 2005-05-31 | 2006-12-07 | Ipflex Inc. | 再構成可能な装置 |
JP4664224B2 (ja) | 2006-03-31 | 2011-04-06 | 富士通セミコンダクター株式会社 | ネットリスト生成方法及び生成装置 |
JP2010277436A (ja) | 2009-05-29 | 2010-12-09 | Renesas Electronics Corp | メモリ構造決定支援装置、メモリ構造決定プログラム及び記録媒体 |
-
2013
- 2013-07-04 US US14/429,157 patent/US9424040B2/en not_active Expired - Fee Related
- 2013-07-04 WO PCT/JP2013/004155 patent/WO2014045500A1/ja active Application Filing
- 2013-07-04 CN CN201380048530.7A patent/CN104641364A/zh active Pending
- 2013-07-04 DE DE112013004631.8T patent/DE112013004631T5/de not_active Withdrawn
- 2013-07-04 JP JP2014536562A patent/JP5835498B2/ja not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH02254496A (ja) * | 1989-03-29 | 1990-10-15 | Yamaha Corp | 楽音発生装置 |
JPH0764957A (ja) * | 1993-08-23 | 1995-03-10 | Mitsubishi Electric Corp | タイマ装置 |
JPH11353265A (ja) * | 1998-06-05 | 1999-12-24 | Mitsubishi Electric Corp | 情報伝達装置 |
JP2004362157A (ja) * | 2003-06-03 | 2004-12-24 | Toshiba Corp | 半導体装置、そのアドレス割り付け方法、および半導体装置の制御方法 |
JP2008097372A (ja) * | 2006-10-12 | 2008-04-24 | Matsushita Electric Ind Co Ltd | システム制御装置 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104597832A (zh) * | 2014-12-31 | 2015-05-06 | 浙江中控研究院有限公司 | 一种基于amba总线的plc程序调度器ip核 |
Also Published As
Publication number | Publication date |
---|---|
US20150234658A1 (en) | 2015-08-20 |
JP5835498B2 (ja) | 2015-12-24 |
JPWO2014045500A1 (ja) | 2016-08-18 |
CN104641364A (zh) | 2015-05-20 |
DE112013004631T5 (de) | 2015-06-25 |
US9424040B2 (en) | 2016-08-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4833303B2 (ja) | 動作合成ツールにおける配列変換 | |
US7325221B1 (en) | Logic system with configurable interface | |
JP2002528825A (ja) | ディジタル信号処理アプリケーション用分散型拡張可能集積回路ディバイスアーキテクチャ | |
US8688947B1 (en) | Aligned data access | |
JP5835498B2 (ja) | Lsi及びlsi製造方法 | |
US20070283260A1 (en) | Human-machine Interface System with Device Bridge and Method for Designing and Operating the Same | |
US20020184471A1 (en) | Semiconductor integrated circuit and computer-readable recording medium | |
KR20170049191A (ko) | 이미지 처리 장치 및 이미지 처리 방법 | |
JP2001043129A (ja) | 半導体装置及びその設計方法 | |
JPWO2006109623A1 (ja) | コンピュータシステム、コンフィギュレーション情報を表すデータ構造、並びにマッピング装置および方法 | |
US7664890B2 (en) | System control device | |
JP2005190487A (ja) | グラフィックスプロセッサ | |
US7996657B2 (en) | Reconfigurable computing circuit | |
JP3476688B2 (ja) | ネットリスト生成方法及びネットリスト生成装置 | |
JP4647578B2 (ja) | レーダ信号処理装置 | |
JP2003274307A (ja) | 多言語サポート対応テレビジョン受像機及びその方法 | |
JP2006309561A (ja) | ダイレクトメモリアクセスコントローラ及びそのデータ転送方法 | |
JP6246585B2 (ja) | 論理回路設計方法及び方法 | |
JP2007323491A (ja) | ダイレクトメモリアクセス制御装置および制御方法 | |
JPH09186836A (ja) | ディジタル複写機 | |
JP2011158983A (ja) | ハードウェア記述言語で記載されたコンピュータプログラム | |
JP2719589B2 (ja) | 1チップ半導体記憶装置 | |
JPH03176738A (ja) | メモリ制御回路 | |
JP2007004424A (ja) | バスシステム | |
JP2006119951A (ja) | バスインターフェース設計支援装置、バスインターフェース設計支援システム、バスインターフェース設計支援プログラムおよびバスインターフェース設計支援方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13838568 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014536562 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14429157 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1120130046318 Country of ref document: DE Ref document number: 112013004631 Country of ref document: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13838568 Country of ref document: EP Kind code of ref document: A1 |