Computer-implemented method, computer-implemented system, and computer-readable medium
By converting the register transmission-level code into a structurally defined circuit design and analyzing it, the power, performance and area optimization problems in integrated circuit design are solved, automated optimization is achieved, development progress is shortened and design efficiency is improved.
Patent Information
- Application Number
- CN202110200470.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-10-21
- Filing Date
- 2021-02-23
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2041-02-23
AI Technical Summary
In the prior art, when optimizing integrated circuit design, especially for RAM and register stack designs of different technical sizes, it is difficult to effectively optimize the power, performance and area of the circuit, and the design process is labor-intensive and time-consuming.
Through the register transmission-level code description of the logical behavior of the input circuit, it is converted into a structurally defined circuit design of multiple types of components and feature size technologies, a plan layout diagram is generated, and the solid analog circuit is analyzed, and the operation conditions are scanned to optimize power , performance and area characteristics.
The automated optimization of integrated circuits is realized, which shortens development progress, reduces redesign work, promotes device verification, and improves design efficiency and accuracy.
Smart Images

Figure CN113836854B_ABST
Abstract
Description
Technical Field
[0001] The techniques described in this disclosure relate generally to integrated circuit system chip design implementation flows, and more particularly to optimized circuit synthesis for implementation on integrated circuits. Background Art
[0002] A variety of techniques may be used to optimize the design of application specific integrated circuits (ASICs) and systems on chips (SoCs) having semiconductor devices. Such designs may be a range of varying silicon designs, implementations, and manufacturing technologies, but are not limited to 3 nanometers, 5 nanometers, 7 nanometers, 10 nanometers, 12 nanometers, 16 nanometers, 28 nanometers, and / or many other technologies. The memory elements of these chips, such as random access memory (RAM), may be optimized based on a variety of techniques using, for example, power performance area. This optimization may be labor intensive and time consuming. Summary of the invention
[0003] The computer-implemented method disclosed herein is used for optimized circuit synthesis implemented on an integrated circuit. The computer-implemented method includes: inputting a register transfer level code description of the logic behavior of the circuit; converting the register transfer level code description into a structurally defined circuit design for multiple types of components and feature size technologies; generating a floor plan of each structurally defined circuit design; creating a physical simulation circuit for each floor plan; and scanning a series of operating conditions to analyze the power, performance, and area of each physical simulation circuit.
[0004] The computer-implemented system disclosed herein is used for optimized circuit synthesis implemented on an integrated circuit. The computer-implemented system includes a conversion module, a simulation module, and an analysis module. The conversion module is configured to access a register transfer level code description of the behavior of the circuit and convert the register transfer level code into a set of structurally defined circuit designs for multiple types of components and feature size technologies. The simulation module is configured to access the set of structurally defined circuit designs and create a physical simulation circuit for each floor plan. The analysis module is configured to access the physical simulation circuit and scan a series of operating conditions of the physical simulation circuit to generate a set of analyses of power, performance, and area characteristics of the physical simulation circuit.
[0005] The non-transitory computer readable medium of the present disclosure encodes a memory storage instruction for manufacturing an integrated circuit. The instructions, when executed, cause operations including: generating a floor plan of each structurally defined circuit design; creating a physical simulation circuit for each floor plan; and scanning a series of operating conditions to analyze the power, performance, and area of each physical simulation circuit. BRIEF DESCRIPTION OF THE DRAWINGS
[0006] The various aspects of the present disclosure will be best understood by reading the following detailed description in conjunction with the accompanying drawings. It should be noted that, in accordance with standard practice in the industry, various features are not drawn to scale. In fact, the size of various features may be arbitrarily increased or reduced for clarity of discussion.
[0007] Figure 1 is a block diagram of an exemplary system for analyzing and extracting data from RAM configurations and register files across various technology sizes (e.g., 5 nm, 7 nm, 10 nm, 12 nm, 16 nm, 28 nm, etc.) according to various embodiments of the present disclosure.
[0008] Figure 2 is a block diagram of another exemplary system for analyzing and extracting data from RAM configurations and register files across various technology sizes (e.g., 5 nm, 7 nm, 10 nm, 12 nm, 16 nm, 28 nm, etc.) in accordance with various embodiments of the present disclosure.
[0009] Figure 3 is a block diagram of another exemplary system for analyzing and extracting data from RAM configurations and register files across various technology sizes (e.g., 5 nm, 7 nm, 10 nm, 12 nm, 16 nm, 28 nm, etc.) in accordance with various embodiments of the present disclosure.
[0010] Figure 4 is an exemplary block diagram illustrating the generation of a synthesized netlist according to various embodiments of the present disclosure.
[0011] Figure 5 is an exemplary block diagram illustrating the generation of a synthesized CPU netlist according to various embodiments of the present disclosure.
[0012] Figure 6 is an exemplary diagram illustrating PPA analysis according to various embodiments of the present disclosure.
[0013] Figure 7 is an exemplary block diagram illustrating the generation of a synthetic ML netlist according to various embodiments of the present disclosure.
[0014] Figure 8 is an exemplary flow chart illustrating a method for optimizing circuit synthesis implemented on an integrated circuit according to various embodiments of the present disclosure.
[0015] Fig. 9 is an exemplary block diagram illustrating a sample computing device architecture for implementing various aspects described herein.
[0016] Description of Figure Numbers
[0017] 100, 200, 300: system;
[0018] 210: conversion module;
[0019] 220: Floor plan mapping module;
[0020] 230: simulation module;
[0021] 240: analysis module;
[0022] 310: classification module;
[0023] 320: display module;
[0024] 330: Circuit design module;
[0025] 340: synthesis module;
[0026] 350: Netlist creation module;
[0027] 400, 500, 700, 900: block diagram;
[0028] 402, 404, 406, 501, 502, 503, 504, 505, 701, 702, 703, 704, 705: file list;
[0029] 410: memory size extractor;
[0030] 422, 424, 426: logic gate configuration;
[0031] 431: Single-port static random access memory;
[0032] 432: Dual-port static random access memory;
[0033] 433: Duplex port static random access memory;
[0034] 434: 2P RF components;
[0035] 436, 520: register compiler;
[0036] 437: 1P RF component;
[0037] 440: wrapper;
[0038] 452, 454, 456: Netlist;
[0039] 460: Block;
[0040] 511, 512, 513, 514: compile;
[0041] 515: register trigger storage;
[0042] 521, 522, 523, 524, 525: CPU netlist;
[0043] 530: User defined;
[0044] 600: illustration;
[0045] 721, 722, 723, 724, 725: ML netlist;
[0046] 800: Flowchart;
[0047] 810, 820, 830, 840, 850: steps;
[0048] 904: bus;
[0049] 908: Processing system;
[0050] 912: read-only memory;
[0051] 914: display interface;
[0052] 916: Random Access Memory;
[0053] 920: Communication port;
[0054] 928: Input device interface;
[0055] 932: Input device;
[0056] 936: keyboard;
[0057] 940: display device;
[0058] 948: disk controller;
[0059] 952, 956: Disk drive. DETAILED DESCRIPTION
[0060] The following disclosure provides many different embodiments or examples for implementing the different features of the provided themes. Specific examples of components and arrangements are described below to simplify the disclosure. Of course, these are only examples and are not intended to be limiting. For example, in the following description, the formation of a first feature above or on a second feature may include an embodiment in which the first feature is formed in direct contact with the second feature, and may also include an embodiment in which an additional feature may be formed between the first feature and the second feature so that the first feature and the second feature may not be in direct contact. In addition, the disclosure may repeat figure numerals and / or letters in various examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or configurations discussed.
[0061] As the size of electronic components continues to decrease and circuit layouts become more complex, simulation of circuit layouts can help optimize circuit design and placement. Thousands of ASIC designs exist for various RAMs, registers, and other storage elements. These designs can be greatly different from each other because each design can consider various types of RAM compilers, cell types (e.g., 6T or 8T), device speed or density, component threshold voltage, power requirements, etc. Compiling these designs together can help optimize new circuit designs. New designs can incorporate appropriate features from existing designs in a way that optimizes the layout for a specific purpose. In other words, the design can be synthesizable for silicon implementation across various device sizes (5 nanometers, 7 nanometers, 10 nanometers, 12 nanometers, 16 nanometers, 28 nanometers, etc.). Compiling a design (e.g., extracted data) includes a description of circuit behavior. Using the extracted data, a new circuit can be physically simulated by translating the circuit behavior into a set of structurally defined circuit layouts that define various electronic component sizes and layouts. The layout can be analyzed to access the physical analog circuit and then a range of operating conditions of the physical analog circuit can be scanned (e.g., using a synthesis variable sweep system (SVSS)) to produce an optimized netlist with a set of analyzed power, performance, and area (PPA) characteristics for the physical analog circuit. This output can provide an optimized physical analog circuit. Such circuits can be applied in many different industries such as high performance computing (HPC), 5G mobile devices, automotive, artificial intelligence or the Internet of Things (IoT).
[0062] Figure 1is a block diagram of an exemplary system 100 for analyzing and extracting data from RAM configurations and register files across various technology sizes (e.g., 5 nm, 7 nm, 10 nm, 12 nm, 16 nm, 28 nm, etc.) according to various embodiments of the present disclosure. System 100 may be a data processing computer system that receives register transfer level (RTL) designs of RAM behavioral models that define the logical behavior of various circuits. RTL is a design abstraction that models synchronous digital circuits in terms of the flow of digital signals between hardware registers and the logical operations performed on those signals. Figures 2 to 3 As described in more detail in , system 100 uses multiple modules to generate an optimized design netlist for a new circuit design. The optimized design netlist has a set of analyses of the power, performance, and area characteristics of a physical analog circuit. A netlist is a description of the connections of an electronic circuit. A netlist is a list that defines the node connections between various electronic components and those components. The new circuit design can be an optimized design based on an RTL design input into system 100. The netlist output by system 100 also includes optimized power, performance, and area (PPA) features. PPA analysis is used to optimize the design because each of the power, performance, and area of a circuit design is an integral aspect of the design. PPA analysis is used during circuit design.
[0063] Figure 2 2 is a block diagram of another exemplary system 200 for analyzing and extracting data from RAM configurations and register files across various technology sizes (e.g., 5 nm, 7 nm, 10 nm, 12 nm, 16 nm, 28 nm, etc.) according to various embodiments of the present disclosure. The system 200 includes a conversion module 210, a floor plan mapping module 220, a simulation module 230, and an analysis module 240. The conversion module 210 receives data such as Figure 1. The register transfer level code is a software code used within a hardware description language. The register transfer level code can be used to describe a physical circuit, which describes the flow of digital signals between hard registers and the logical operations (e.g., AND, NAND, OR, NOR) performed on the digital signals by the components of the physical circuit. The conversion module 210 accesses the register transfer level code description of the circuit logic behavior. The conversion model 210 then converts the register transfer level code into a set of structurally defined circuit designs (e.g., AND gates, NAND circuits, OR gates, NOR circuits) and feature size technologies (e.g., 5 nanometers, 7 nanometers, 10 nanometers, 12 nanometers, 16 nanometers, 28 nanometers, etc.) for multiple types of components (e.g., CPUs, GPUs, ML-NPUs). The floor plan mapping module 220 accesses a set of structurally defined circuit designs and converts the set of structurally defined circuit designs into a set of floor plans. The floor plan defines the electronic components within the circuit design and the physical placement of the components in a specific area (e.g., bottom layer). The simulation module 230 accesses a set of floor plans and creates a simulatable circuit for each floor plan. The analysis module 240 accesses the physical simulation circuit and scans a series of operating conditions of the physical simulation circuit to generate an optimized design netlist with a set of analyses of power, performance, and area characteristics of the physical simulation circuit. The analysis module 240 also scans a series of operating conditions (e.g., various input voltages, various environments such as high temperature / low temperature, high humidity / low humidity, etc.) for multiple physical simulation circuits simultaneously in parallel (e.g., simultaneously).
[0064] Figure 3 is a block diagram of another exemplary system 300 for analyzing and extracting data from RAM configurations and register files across various technology sizes (e.g., 5 nm, 7 nm, 10 nm, 12 nm, 16 nm, 28 nm, etc.) according to various embodiments of the present disclosure. Figure 2 Similar to system 200, system 300 may include conversion module 210, floorplan mapping module 220, simulation module 230, and analysis module 240. In addition, system 300 may include one or more of classification module 310, display module 320, circuit design module 330, synthesis module 340, and netlist creation module 350.
[0065] The classification module 310 accesses the analysis and generates a classification list of a set of physical analog circuits through power, performance and area characteristics, which define the power consumption, performance (that is, the frequency of circuit operation) and the area occupied by the circuit. These characteristics can be classified based on power, performance and area values. The display module 320 accesses the classification list of physical analog circuits and displays the classification list of physical analog circuits on a graphical user interface. The circuit design module 330 accesses a set of structurally defined circuit designs and converts the structurally defined circuit designs into a set of file lists (filelist) for the synthesis of multiple profiled configurations of power, performance and area characteristics. The file list defines the components of the circuit and the interconnection points between the components. The synthesis module 340 accesses a set of file lists and synthesizes a set of file lists into a set of logic gate configurations using a design compiler. The synthesis of the circuit includes translating the register transfer level code into a general Boolean model. The Boolean model can be optimized at the logic level to combine multiple operations into simpler operations. For example, components may be rearranged, combined, and / or recombined into different components in an effort to reduce the operation time and the number of overall components within the circuit. Additionally, optimization may include a sequential mapping of components, where logic surrounding sequential cells may be absorbed within a cell to save area. The design compiler may also perform various optimization operations, such as removing logic that does not have output port connections, removing redundant components (e.g., such as two sequential inverters), and / or propagating constants to reduce the overall logic components within the circuit. The netlist creation module 350 accesses a set of logic gate configurations and converts a set of logic gate configurations into a set of netlists to be used by the simulation module 230. The logic gate configurations contain a logic schematic that defines the interconnections of the components. These logic gate configurations may be translated into a textual representation of the nodes within the configuration, the components within the configuration, and their interconnections. This textual representation is a netlist.
[0066] Figure 4 4 is an exemplary block diagram 400 illustrating the generation of a synthesized netlist according to various embodiments of the present disclosure. The file list (e.g., RTL design describing the logic behavior of the circuit - software code defining the logic behavior of the circuit) may involve a variety of technologies, such as a central processing unit (CPU) (e.g., CPU-IP1402), a graphical processing unit (GPU) (e.g., GPU-IP1), a machine learning network processor (ML-NPU) (e.g., ML-NPU-IP1), etc. Each set of file lists may include one or more file lists associated with different module designs, such as Figure 6 to Figure 7 4. The memory size extractor 410 (e.g., synthesis module 340) processes a given group netlist to generate a logic gate configuration corresponding to each group, the logic gate configuration defining the size of a given memory cell being evaluated. For example, a logic gate configuration 422 (e.g., config.CPU) corresponds to the CPU-IP1 file list 402. The memory size extractor 410 also includes a tracker. The tracker generates behavioral information associated with at least a portion of an RTL design corresponding to a semiconductor device. For example, this behavioral information may include device size (e.g., the number of word lines and the number of bit lines). Similarly, a logic gate configuration 424 (e.g., config.GPU) corresponds to the GPU-IP1 file list 404. A logic gate configuration 426 (e.g., config.ML-NPU) corresponds to the ML-NPU-IP1 file list 406. As previously described in detail, logic gate configurations are compiled using register compilers such as single-port (SP) static random access memory (SRAM) 431, dual-port SRAM 432, dual-port (DP) SRAM 433, 1PRF components 437, 2P RF components 434, and any other type of register compiler 436. These compiled configurations are then translated into netlists corresponding to input file lists (e.g., file list 402, file list 404, file list 406) by wrappers 440. A wrapper is software defined to encapsulate the wrapped code so that the code complexity of the wrapped code is not visible to another interfacing software component. In fact, the wrapper itself contains simpler code that creates an interface with the other software component to simplify the interaction between the two software components. For example, two software components can use wrappers when they have different types of software code that are incompatible with each other. For example, netlist CPU-IP1452 corresponds to file list CPU-IP1402. Netlist GPU-IP 2454 corresponds to file list GPU-IP 1404. Netlist MP-NPU-IP 1456 corresponds to ML-NPU-IP 1406. These netlists are used to generate optimized circuit designs (e.g., block 460) using automatic placement of components and / or routing of signals. The optimized netlists are transmitted back to the compiler and stored so that they can be used in future circuit layouts.
[0067] Figure 5500 is an exemplary block diagram illustrating the generation of a synthesized CPU netlist according to various embodiments of the present disclosure. In this example, a set of file lists corresponds to N modules (e.g., file list 501, file list 502, file list 503, file list 504, file list 505, etc.). Each file list contains specific behavioral logic associated with the specific module. A set of file lists are compiled together into groupings based on technology size (e.g., 5 nm, 7 nm, 16 nm, etc.) using a register compiler or a trigger compiler 520. The compilation is used to generate a layout for an SRAM block in a layout. Each compilation includes all physical memory configurations and PPA configurations associated with input file list 501, file list 502, file list 503, file list 504, file list 505. In other words, all modules with one word line and one bit line (e.g., module-1 of file list 502) across various technology sizes (e.g., 5 nm, 7 nm, 16 nm, etc.) are compiled together. Similarly, all modules with two word lines and two bit lines (e.g., module-2 of file list 052) across various technology sizes (e.g., 5 nm, 7 nm, 16 nm, etc.) are compiled together. This operation continues for all modules across all various technology sizes (e.g., 5 nm, 7 nm, 16 nm, etc.). The PPA profiler evaluates the stored compilations (e.g., compilation 511, compilation 512, compilation 513, compilation 514) to logic gate configurations for each technology size (e.g., 5 nm, 7 nm, 16 nm, etc.). Figure 6 PPA analysis is described in more detail in . Register compiler 520 settings may be stored in register trigger storage 515. Compile 511, compile 512, compile 513, compile 514 along with register trigger storage 515 may be wrapped and compiled to generate corresponding netlists (e.g., CPU netlist 521, CPU netlist 522, CPU netlist 523, CPU netlist 524, CPU netlist 525) with wrapped PPA profiles that define power, performance, and area characteristics of various modules. A wrapper is software defined to encapsulate wrapped code so that the code complexity of the wrapped code is not visible to another interfacing software component. In fact, the wrapper itself contains simpler code that creates an interface with the other software component thereby simplifying the interaction between the two software components. For example, a wrapper may be used when two software components have different types of software code that are incompatible with each other. In Figure 5 In the example of the embodiment shown in , the PPA profile defines specific power, performance, and area requirements associated with different module sizes (e.g., one wordline / bitline, two wordlines / bitlines, etc.). In some embodiments, a project user-defined configurator may set a specific size definition (e.g., user definition 530) for the circuit to be optimized. In other words, the user may set a specific size for the circuit via user input.
[0068] Figure 6 600 is an exemplary diagram illustrating PPA analysis. Various file lists are evaluated using the PPA analyzer to simultaneously increase performance, maintain low power consumption, and minimize device area. These calculations are performed by triangulating constants associated with power (e.g., sp, ip), area (e.g., udh), and performance (e.g., hs). These constants define the power consumed by an electronic device, the area occupied by electronic device components, and the clock frequency (e.g., performance) required to operate the electronic device. Triangulation is a process of measuring the distance between each of the constants (e.g., the points shown on diagram 600). The ultimate goal is to make the power constant sp, the power constant ip, and the area constant udh at their lowest possible values without sacrificing circuit performance hs. All of these three metrics (e.g., power, area, performance) are interdependent, so modifying one of these values affects the other two.
[0069] Figure 7 is an exemplary block diagram 700 illustrating the generation of a synthetic ML netlist according to various embodiments of the present disclosure. In this example, a set of file lists corresponds to N modules (e.g., file list 701, file list 702, file list 703, file list 704, file list 705, etc.). Each file list contains specific behavioral logic associated with the particular module. File list 701 describes a first module (e.g., module-1) having one word line and one bit line. File list 702 describes a second module (e.g., module-2) having two word lines and two bit lines. File list 703 describes a third module (e.g., module-3) having three word lines and three bit lines. File list 704 describes a smaller RAM having at least four word lines and at least four bit lines. File list 705 describes an Nth module having any number (e.g., N) of word lines and any number (e.g., N) of bit lines. As shown in Figure 7 As described in detail in , file list 701, file list 702, file list 703, file list 704, file list 705 are processed and corresponding netlists (e.g., ML netlist 721, ML netlist 722, ML netlist 723, ML netlist 724, ML netlist 725) with packaged PPA profiles are generated.
[0070] Figure 8 is an exemplary flow chart 800 illustrating a method for optimizing circuit synthesis implemented on an integrated circuit according to various embodiments of the present disclosure. Although reference is made to the previously described structural description herein for ease of understanding, Figure 8, but it should be understood that the method is also applicable to many other structures. A register transfer level code description of the logical behavior of the circuit is input through the conversion module 210 (e.g., step 810). The register transfer level code description is converted into a structurally defined circuit design for multiple types of components and feature size technologies using the conversion module 210 (e.g., step 820). A floor plan mapping module 220 is used to generate a floor plan of each structurally defined circuit design based on the structural design circuit (e.g., step 830). A physical simulation circuit is created for each floor plan to generate the physical simulation circuit using the simulation module 230 (e.g., step 840). A series of operating conditions are scanned to analyze the power, performance, and area of each physical simulation circuit using the analysis module 240 (e.g., step 850).
[0071] Fig. 9 900 is an exemplary block diagram illustrating a sample computing device architecture for implementing various aspects described herein. Bus 904 may serve as an information highway interconnecting other illustrated components of hardware. A processing system 908 (e.g., one or more computer processors / data processors at a given computer or multiple computers), labeled as a central processing unit (CPU), may perform the computational and logical operations required to execute a program. Non-transitory processor-readable storage media, such as read only memory (ROM) 912 and random access memory (RAM) 916 may communicate with the processing system 908 and may contain one or more programming instructions for the operations specified herein. Optionally, the program instructions may be stored on a non-transitory computer-readable storage medium, such as a magnetic disk, an optical disk, a recordable storage device, a flash memory, or other physical storage media.
[0072] In one example, a disk controller 948 may interface one or more optional disk drives to the system bus 904. These disk drives may be external or internal CD-ROM, CD-R, CD-RW, or DVD, or a solid-state drive such as 952, or an external or internal hard drive 956. As previously indicated, these various disk drives 952, disk drives 956, and disk controllers are optional devices. The system bus 904 may also include at least one communication port 920 to allow communication with external devices that are physically connected to the computing system or that may be obtained externally via a wired or wireless network. In some cases, the communication port 920 includes or otherwise comprises a network interface.
[0073] To provide interaction with a user, the subject matter described herein may be implemented on a computing device having a display device 940 (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information obtained from the bus 904 to the user and an input device 932, such as a keyboard 936 and / or a pointing device (e.g., a mouse or navigation ball) and / or a touch screen, by which the user can provide input to the computer. Other types of input devices 932 may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback through a microphone, or tactile feedback); and the input from the user may be received in any form, including sound waves, voice, or tactile input. The input device 932 and the keyboard 936 may be coupled to the bus 904 by means of an input device interface 928 and transmit information via the bus 904. Other computing devices, such as dedicated servers, may omit one or more of the display 940 and display interface 914 , the input device 932 , the keyboard 936 , and the input device interface 928 .
[0074] In addition, the methods and systems described herein may be implemented on many different types of processing devices by program code including program instructions executable by a device processing subsystem. The software program instructions may include source code, object code, machine code, or any other stored data operable to cause a processing system to perform the methods and operations described herein, and may be provided in any suitable language such as C, C++, JAVA, Perl, Python, Tcl, or any other suitable programming language. However, other implementations may also be used, such as firmware or even appropriately designed hardware configured to perform the methods and systems described herein.
[0075] System data and method data (e.g., associations, mappings, data inputs, data outputs, intermediate data results, final data results, etc.) may be stored and implemented in one or more different types of computer-implemented data storage, such as different types of storage devices and program constructs (e.g., RAM, ROM, flash memory, flat files, databases, program data structures, program variables, IF-THEN (or similar type) statement constructs, etc.) It should be noted that data structures describe formats for organizing and storing data in a database, program, memory, or other computer-readable medium used by a computer program.
[0076] The computer components, software modules, functions, data storage devices, and data structures described herein may be connected to each other directly or indirectly to allow data to flow as required for their operation. It should also be noted that a module or processor includes but is not limited to a code unit that performs software operations, and may be implemented, for example, as a subroutine unit of code, or as a software function unit of code, or as an object (such as in an object-oriented paradigm table), or as an applet, or implemented in a computer scripting language, or as another type of computer code. Software components and / or functionality may be located on a single computer or distributed across multiple computers depending on the exigencies.
[0077] The use of the various processes described herein can provide several advantages. For example, the use of the subject matter can shorten the ASIC development schedule because the circuit is automatically optimized by design and minimal redesign effort is required. The simulation described herein also facilitates device verification without the time delays associated with producing and manually testing devices.
[0078] In one embodiment, a method for optimizing circuit synthesis implemented on an integrated circuit includes inputting a register transfer level code description of the logic behavior of the circuit. Converting the register transfer level code description into structurally defined circuit designs for multiple types of components and feature size technologies. Generating floor plans for each structurally defined circuit design. Creating a physical simulation circuit for each floor plan. Scanning a range of operating conditions to analyze the power, performance, and area of each physical simulation circuit.
[0079] In some embodiments, the computer-implemented method further includes classifying the physical analog circuits by power, performance, and area characteristics.
[0080] In some embodiments, the computer-implemented method further includes displaying the physical simulation circuits categorized by power, performance, and area characteristics on a graphical user interface.
[0081] In some embodiments, the computer-implemented method further includes generating a list of files for synthesis of multi-profile configurations of power, performance, and area characteristics prior to generating a floorplan of each structurally defined circuit design.
[0082] In some embodiments, the computer-implemented method further includes synthesizing the file list into a logic gate configuration using a design compiler.
[0083] In some embodiments, the computer-implemented method further includes generating a netlist for each logic gate configuration.
[0084] In some embodiments, the computer-implemented method of scanning a range of operating conditions is performed simultaneously in parallel on multiple physical analog circuits.
[0085] In another embodiment, a computer-implemented system for optimizing circuit synthesis implemented on an integrated circuit includes a conversion module, a floorplan mapping module, a simulation module, and an analysis module. The conversion module is configured to access a register transfer level code description of the behavior of the circuit and convert the register transfer level code into a set of structurally defined circuit designs for multiple types of components and feature size technologies. The simulation module is configured to access a set of structurally defined circuit designs and create a physical simulation circuit for each floorplan. The analysis module is configured to access the physical simulation circuit and scan a series of operating states of the physical simulation circuit to generate a set of analyses of power, performance, and area characteristics of the physical simulation circuit.
[0086] In some embodiments, the computer-implemented system further comprises a classification module configured to access the analysis and generate a classification list of the set of physical analog circuits according to power, performance, and area characteristics.
[0087] In some embodiments, the computer-implemented system further comprises a display module configured to access the classified list of physical analog circuits and display the classified list of physical analog circuits on a graphical user interface.
[0088] In some embodiments, the computer-implemented system further includes a circuit design module. The circuit design module is configured to access the set of structurally defined circuit designs and convert the structurally defined circuit designs into a set of file lists for synthesis of multi-profile configurations for power, performance, and area characteristics.
[0089] In some embodiments, the computer-implemented system further comprises a synthesis module configured to access the set of file lists and synthesize the set of file lists into a set of logic gate configurations using a design compiler.
[0090] In some embodiments, the computer-implemented system further comprises a netlist creation module configured to access the set of logic gate configurations and convert the set of logic gate configurations into a set of netlists.
[0091] In some embodiments, the analysis module simultaneously scans a series of operating conditions on multiple physical analog circuits in parallel.
[0092] In yet another embodiment, a non-transitory computer readable medium is encoded with memory storing instructions for fabricating an integrated circuit, the instructions when executed causing various operations, such as generating a floor plan of each structurally defined circuit design. Creating a physical analog circuit for each floor plan. Scanning a range of operating conditions to analyze power, performance, and area of each physical analog circuit.
[0093] In some embodiments, a non-transitory computer-readable medium includes operations for classifying physical analog circuits by power, performance, and area characteristics.
[0094] In some embodiments, the non-transitory computer-readable medium includes operations for generating a netlist for synthesis of multi-profile configurations of power, performance, and area characteristics prior to generating a floorplan of each structurally defined circuit design.
[0095] In some embodiments, the non-transitory computer-readable medium includes operations for synthesizing the list of files into a logic gate configuration using a design compiler.
[0096] In some embodiments, the non-transitory computer-readable medium includes operations for generating a netlist for each logic gate configuration.
[0097] In some embodiments, the scanning of a series of operating conditions is performed simultaneously in parallel on multiple physical analog circuits.
[0098] The foregoing summarizes the features of several embodiments so that those skilled in the art can better understand the various aspects of the present disclosure. Those skilled in the art will appreciate that they can easily use the present disclosure as a basis for designing or modifying other processes and structures for achieving the same purposes and / or achieving the same advantages of the embodiments introduced herein. Those skilled in the art will also recognize that such equivalent constructions do not depart from the spirit and scope of the present disclosure, and that they may make various changes, substitutions, and modifications herein without departing from the spirit and scope of the present disclosure.
Claims
1. A computer-implemented method for optimizing circuit synthesis implemented on an integrated circuit, characterized in that The computer-implemented method comprises: A register transfer level code description of the logic behavior of the input circuit; converting the register transfer level code description into a structurally defined circuit design for various types of components and feature size technologies; generating a floor plan of each structurally defined circuit design; Creating a physical simulation circuit for each floor plan; Sweep a range of operating conditions to analyze the power, performance, and area of each physical analog circuit; classifying the physical analog circuits by power, performance, and area characteristics; and The graphical user interface displays the physical simulation circuits sorted by power, performance, and area characteristics.
2. The computer-implemented method of claim 1, wherein: Further included is generating a list of files for synthesis of multiple profiles of power, performance, and area characteristics prior to generating a floor plan of each structurally defined circuit design.
3. The computer-implemented method of claim 2, wherein: The method further includes synthesizing the file list into a logic gate configuration using a design compiler.
4. The computer-implemented method of claim 3, wherein: The method further includes generating a netlist of each logic gate configuration.
5. The computer-implemented method of claim 1, wherein: The scanning of a series of operating conditions is performed simultaneously and in parallel on a plurality of physical analog circuits.
6. A computer-implemented system for optimizing circuit synthesis implemented on an integrated circuit, characterized in that The computer-implemented system comprises: a conversion module configured to access a register transfer level code description of the behavior of the circuit and convert the register transfer level code into a set of architecturally defined circuit designs for multiple types of components and feature size technologies; a simulation module configured to access the set of structurally defined circuit designs and create a physical simulation circuit for each floor plan; an analysis module configured to access the physical analog circuit and scan a series of operating conditions of the physical analog circuit to generate a set of analyses of power, performance, and area characteristics of the physical analog circuit; a classification module configured to access the analysis and generate a classification list of the set of physical analog circuits by power, performance, and area characteristics; and The display module is configured to access the classification list of physical analog circuits and display the classification list of physical analog circuits on a graphical user interface.
7. The computer-implemented system of claim 6, wherein: The computer-implemented system further includes a circuit design module configured to access the set of architecturally defined circuit designs and convert the architecturally defined circuit designs into a set of file lists for synthesis of multiple profiles of power, performance, and area characteristics.
8. The computer-implemented system of claim 7, wherein: The computer-implemented system further includes a synthesis module configured to access the set of file lists and synthesize the set of file lists into a set of logic gate configurations using a design compiler.
9. The computer-implemented system of claim 8, wherein: The computer-implemented system further includes a netlist creation module configured to access the set of logic gate configurations and convert the set of logic gate configurations into a set of netlists.
10. The computer-implemented system of claim 6, wherein: The analysis module simultaneously and in parallel scans a series of operating conditions for multiple physical analog circuits.
11. A non-transitory computer readable medium encoding memory storage instructions for manufacturing an integrated circuit, characterized in that: The instructions, when executed, result in operations including: generating a floor plan of each of a plurality of structurally defined circuit designs; Creating a physical simulation circuit for each floor plan; and Sweep a range of operating conditions to analyze the power, performance, and area of each physical analog circuit; classifying the physical analog circuits by power, performance, and area characteristics; and Displays a physical analog circuit sorted by power, performance, and area characteristics.
12. The non-transitory computer-readable medium of claim 11, wherein: Further included is generating a netlist for synthesis of power, performance, and area characteristics prior to generating a floorplan of each structurally defined circuit design.
13. The non-transitory computer readable medium of claim 12, further comprising synthesizing the file list into a logic gate configuration using a design compiler.
14. The non-transitory computer readable medium of claim 13, wherein: The method further includes generating a netlist of each logic gate configuration.
15. The non-transitory computer readable medium of claim 11, wherein: The scanning of a series of operating conditions is performed simultaneously and in parallel on a plurality of physical analog circuits.
