CN109840878B - GPU parameter management method based on SystemC - Google Patents
GPU parameter management method based on SystemC Download PDFInfo
- Publication number
- CN109840878B CN109840878B CN201811520556.8A CN201811520556A CN109840878B CN 109840878 B CN109840878 B CN 109840878B CN 201811520556 A CN201811520556 A CN 201811520556A CN 109840878 B CN109840878 B CN 109840878B
- Authority
- CN
- China
- Prior art keywords
- register
- executing
- graphic
- module
- level interface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Image Generation (AREA)
Abstract
The invention relates to the technical field of computer hardware modeling, and provides a GPU-oriented parameter management method based on SystemC, which comprises the following steps: step 1: performing barrier/lock management, calculating a new barrier state according to the current barrier state and the FIFO state, judging that when the graphRegLock is 0, outputting the barrier state to a graph management module (2) through a transaction level interface, executing step 2, otherwise, when the graphdraglock is 0 and the graphFunLock is 0, outputting the barrier state to a register window management module (3) through the transaction level interface, and executing step 3; step 2: executing graphic drawing command parameters and function code parameter management; step 3: read-write management of register parameters is performed.
Description
Technical Field
The invention relates to the technical field of computer hardware modeling, in particular to a GPU-oriented parameter management method based on SystemC.
Background
With the increasing number of graphics applications, early solutions for graphics rendering by CPU alone have been difficult to meet the ever-increasing graphics processing demands of performance and technology, and graphics processors (Graphic Processing Unit, GPU) have grown. The first GPU product released in Nvidia in 1999 has been developed by GPU technology, which mainly goes through a fixed function pipeline stage, a separate shader architecture stage, and a unified shader architecture stage, so that the graphics processing capability is continuously improved, and the application field is gradually expanded from the initial graphics drawing to the general computing field. The GPU pipeline has high speed, parallel characteristics and flexible programmable capability, and provides a good operation platform for graphic processing and general parallel computing.
The development of the GPU chip has huge hardware logic scale and higher complexity, and the design needs to be described at a higher level of abstraction so as to perform higher-speed simulation, software/hardware co-simulation and exploration of a system architecture. When the design is expressed as a system level model, multiple attempts to design by using different algorithms are easy to achieve, and experiments can be completed quickly by changing different structures; if the design is expressed using a register transfer level or gate level model, the scale is typically quite large, and it is time consuming and laborious to try out a different design structure or make some changes, if not too difficult.
The key factor of SystemC as a language to promote its development and standardization is that system level design can be performed, and the architecture of hardware and the algorithm of software can be described, supporting verification and communication of IP. The use of SystemC as a partitioning tradeoff for software and hardware is much easier than other languages at the system level and simulation is much faster than using multiple languages. The use of a microstructure based on SystemC to design and describe the cells thus enables a fully standard simulation environment to be built, modeling directly at a high level of abstraction.
Disclosure of Invention
Based on the problems in the background art, the GPU parameter management method based on the SystemC can solve the problem of accurate data comparison of the RTL simulation GPU parameter management unit, and can verify functions of the hardware microstructure of the GPU parameter management unit on the TLM model in advance.
The technical scheme of the invention is as follows:
a GPU-oriented parameter management method based on SystemC comprises the following steps:
step 1: performing barrier/lock management, calculating a new barrier state according to the current barrier state and the FIFO state, judging that when the graphRegLock is 0, outputting the barrier state to the graphic management module 2 through a transaction level interface, executing step 2, otherwise, when the graphdraglock is 0 and the graphFunLock is 0, outputting the barrier state to the register window management module 3 through the transaction level interface, and executing step 3;
step 2: executing graphic drawing command parameters and function code parameter management;
step 3: read-write management of register parameters is performed.
The step 1 comprises the following steps:
step 11: setting graphfurbar to 1 when graphDrawBar is 0, graphfurnbar is 0, graphdrawfifostatus is non-empty, and graphfurfifostatus is empty; otherwise, executing the step 12;
step 12: setting graphDrawBar to 1 when graphDrawBar is 0, graphfurnbar is 0, graphdrawfafostatus is empty, and graphfurffafostatus is non-empty; otherwise, executing the step 13;
step 13: setting the graphdraw bar to 1 and the graphfunbar to 0 when the graphdraw bar is 0, the graphfunbar is 1, the graphdraw fifo status is empty and the graphfunfifo status is not empty; otherwise, executing the step 14;
step 14: when the graphDrawBar is 1, the graphfurnbar is 0, the graphdrawfafofstatus is non-empty and the graphfurafofstatus is empty, setting the graphDrawBar to 0 and the graphfurnbar to 1; otherwise, executing the step 15;
step 15: reading graphDrawLock, graphFunLock, graphRegLock through the transaction level interface, executing step 16 when the graphRegLock is 0, and executing step 17 when the graphdraglock is 0 and the graphFunLock is 0;
step 16: outputting all variable information and FIFO data information to a graphic management module 2 through a transaction level interface, and executing graphic parameter management;
step 17: all the variable information and the register window data information are output to the register window management module 3 through the object level interface, and register parameter management is executed.
The step 2 comprises the following steps:
step 21: the graph detection module 21 reads the variable graphRegLock from the barrier/lock management module 1, and if the variable graphRegLock is a lock, the step 21 is looped; otherwise, executing step 22;
step 22: the graphics detection module 21 reads the variable graphDrawBar, graphFunBar, graphDrawFifoStatus, graphFunFifoStatus from the barrier/lock management module 1, and when the graphics drawbar is 0, the graphics funnbar is 1, and the graphics drawfifostatus is non-empty, it is connected to the drawing command processing module 22 through the transaction-level interface, and step 23 is executed; otherwise, when the graphdraw bar is 1, the graphfunbar is 0, and the graphFunFifoStatus is not empty, the step 24 is executed by connecting to the function code processing module 23 through the transaction level interface; otherwise, returning to the step 21;
step 23: the drawing command processing module 22 firstly sets the graphdragwlock to 1, then decodes and assembles the drawing parameters of the graph, and sends the assembled data to the task scheduling module 5 through the object-level interface, and sets the graphdragwlock to 0 after the execution is finished;
step 24: the function code processing module 23 firstly sets 1 to the graphFunLock, then decodes and assembles the graphic function parameters, and sends the assembled data to the corresponding drawing module 6 through the object-level interface, and sets 0 to the graphFunLock after execution.
The step 3 comprises the following steps:
step 31: setting graphRegLock to 1;
step 32: judging the register read-write type DataType, if the register read-write type DataType is read, executing the step 33; if the write is made, go to step 34;
step 33: connected to the external register module 7 through the transaction level interface, performing writing of corresponding register data DataStruct according to the register address DataAddr;
step 34: connected to the external register module 7 through the transaction-level interface, performs reading out the corresponding register data DataStruct according to the register address DataAddr;
step 35: the graphRegLock is set to 0.
The method comprises the following steps:
internally two FIFOs are included, receiving two types of commands from the command processing module 4: wherein the drawing commands are stored in a graphics drawing command FIFO and the graphics function commands are stored in a graphics function command FIFO;
the data information is contained in the data information: the register window data specifically comprises a register read-write type DataType, register data DataStruct and a register address DataAddr;
the internal contains the variables:
the variable used to represent the barrier state of the graphics drawing command, denoted as graphdragbus, is initialized to 0;
the variable used to represent the status of the graphics function command barrier, noted graphFunBar, is initialized to 0;
the variable used to represent the status of the graphics drawing command FIFO is denoted as graphDrawFifoStatus;
the variable used to represent the status of the graphics function command FIFO is denoted as graphFunFifoStatus;
the variable used to represent the graphics drawing command lock state is denoted as graphdraglock;
the variable used for representing the command lock state of the graphic function is recorded as graphic FunLock;
the variable used to represent the status of the register window lock is denoted graphRegLock.
The beneficial effects of the invention are as follows:
1. according to the GPU parameter management method based on SystemC, which is provided by the invention, the barrier/lock management module 1, the graphic management module 2 and the register window management module 3 are integrated inside, so that the distribution and management of various parameters of a GPU can be realized.
2. The barrier/lock management module 1 subunit realizes the monitoring and management of locks and barriers, and ensures the sequential process of various parameters in the management process.
3. The graphic management module 2 subunit realizes the distribution and management functions of graphic drawing parameters, and specifically comprises two types of parameter management of graphic drawing parameters and graphic function parameters.
4. The register window management module 3 subunit implements the window register allocation and management functions.
5. The method solves the problem of comparison of RTL simulation result models of the GPU parameter management unit, solves the problem of function verification of the TLM microstructure unit facing the hardware of the GPU parameter management unit, and accelerates the simulation speed.
Drawings
FIG. 1 is a block diagram of a GPU-oriented parameter management method;
fig. 2 is a TLM microstructure block diagram of the graphics management module 2 hardware.
Detailed Description
A GPU-oriented parameter management method based on SystemC comprises the following steps:
step 1: performing barrier/lock management, calculating a new barrier state according to the current barrier state and the FIFO state, judging that when the graphRegLock is 0, outputting the barrier state to the graphic management module 2 through a transaction level interface, executing step 2, otherwise, when the graphdraglock is 0 and the graphFunLock is 0, outputting the barrier state to the register window management module 3 through the transaction level interface, and executing step 3;
step 2: executing graphic drawing command parameters and function code parameter management;
step 3: read-write management of register parameters is performed.
The step 1 comprises the following steps:
step 11: setting graphfurbar to 1 when graphDrawBar is 0, graphfurnbar is 0, graphdrawfifostatus is non-empty, and graphfurfifostatus is empty; otherwise, executing the step 12;
step 12: setting graphDrawBar to 1 when graphDrawBar is 0, graphfurnbar is 0, graphdrawfafostatus is empty, and graphfurffafostatus is non-empty; otherwise, executing the step 13;
step 13: setting the graphdraw bar to 1 and the graphfunbar to 0 when the graphdraw bar is 0, the graphfunbar is 1, the graphdraw fifo status is empty and the graphfunfifo status is not empty; otherwise, executing the step 14;
step 14: when the graphDrawBar is 1, the graphfurnbar is 0, the graphdrawfafofstatus is non-empty and the graphfurafofstatus is empty, setting the graphDrawBar to 0 and the graphfurnbar to 1; otherwise, executing the step 15;
step 15: reading graphDrawLock, graphFunLock, graphRegLock through the transaction level interface, executing step 16 when the graphRegLock is 0, and executing step 17 when the graphdraglock is 0 and the graphFunLock is 0;
step 16: outputting all variable information and FIFO data information to a graphic management module 2 through a transaction level interface, and executing graphic parameter management;
step 17: all the variable information and the register window data information are output to the register window management module 3 through the object level interface, and register parameter management is executed.
The step 2 comprises the following steps:
step 21: the graph detection module 21 reads the variable graphRegLock from the barrier/lock management module 1, and if the variable graphRegLock is a lock, the step 21 is looped; otherwise, executing step 22;
step 22: the graphics detection module 21 reads the variable graphDrawBar, graphFunBar, graphDrawFifoStatus, graphFunFifoStatus from the barrier/lock management module 1, and when the graphics drawbar is 0, the graphics funnbar is 1, and the graphics drawfifostatus is non-empty, it is connected to the drawing command processing module 22 through the transaction-level interface, and step 23 is executed; otherwise, when the graphdraw bar is 1, the graphfunbar is 0, and the graphFunFifoStatus is not empty, the step 24 is executed by connecting to the function code processing module 23 through the transaction level interface; otherwise, returning to the step 21;
step 23: the drawing command processing module 22 firstly sets the graphdragwlock to 1, then decodes and assembles the drawing parameters of the graph, and sends the assembled data to the task scheduling module 5 through the object-level interface, and sets the graphdragwlock to 0 after the execution is finished;
step 24: the function code processing module 23 firstly sets 1 to the graphFunLock, then decodes and assembles the graphic function parameters, and sends the assembled data to the corresponding drawing module 6 through the object-level interface, and sets 0 to the graphFunLock after execution.
As shown in fig. 2.
The step 3 comprises the following steps:
step 31: setting graphRegLock to 1;
step 32: judging the register read-write type DataType, if the register read-write type DataType is read, executing the step 33; if the write is made, go to step 34;
step 33: connected to the external register module 7 through the transaction level interface, performing writing of corresponding register data DataStruct according to the register address DataAddr;
step 34: connected to the external register module 7 through the transaction-level interface, performs reading out the corresponding register data DataStruct according to the register address DataAddr;
step 35: the graphRegLock is set to 0.
The method comprises the following steps:
internally two FIFOs are included, receiving two types of commands from the command processing module 4: wherein the drawing commands are stored in a graphics drawing command FIFO and the graphics function commands are stored in a graphics function command FIFO;
the data information is contained in the data information: the register window data specifically comprises a register read-write type DataType, register data DataStruct and a register address DataAddr;
the internal contains the variables:
the variable used to represent the barrier state of the graphics drawing command, denoted as graphdragbus, is initialized to 0;
the variable used to represent the status of the graphics function command barrier, noted graphFunBar, is initialized to 0;
the variable used to represent the status of the graphics drawing command FIFO is denoted as graphDrawFifoStatus;
the variable used to represent the status of the graphics function command FIFO is denoted as graphFunFifoStatus;
the variable used to represent the graphics drawing command lock state is denoted as graphdraglock;
the variable used for representing the command lock state of the graphic function is recorded as graphic FunLock;
the variable used to represent the status of the register window lock is denoted graphRegLock.
The GPU parameter management method based on SystemC comprises a barrier/lock management module 1, a graphic management module 2 and a register window management module 3.
The barrier/lock management module 1 is connected to an external command processing module 4 through a transaction level interface; the two subunits of the graphic management module 2 and the register window management module 3 are mutually independent in physical and logical aspects and are connected with the barrier gate/lock management module 1 through a transaction-level interface; the graphic management module 2 is connected to the external task scheduling module 5 and the drawing module 6 through a transaction-level interface; the register window management module 3 is connected to an external register module 7 via a transaction level interface.
As shown in fig. 1.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solution of the present invention, and not limiting thereof; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (5)
1. A GPU-oriented parameter management method based on SystemC is characterized by comprising the following steps: comprising the following steps:
step 1: performing barrier/lock management, calculating a new barrier state according to the FIFO state of the graphic drawing command and the FIFO state of the graphic function command, judging that when the graphic reglock is 0, outputting all variable information and FIFO data information to a graphic management module (2) through a transaction level interface, executing step 2, otherwise, when the graphic draglock is 0 and the graphic funlock is 0, outputting all variable information and register window data information to a register window management module (3) through the transaction level interface, and executing step 3;
the variable used to represent the graphics drawing command lock state is denoted as graphdraglock;
the variable used for representing the command lock state of the graphic function is recorded as graphic FunLock;
the variable used for representing the lock state of the register window is marked as graphRegLock;
step 2: executing graphic drawing command parameters and function code parameter management;
step 3: read-write management of register parameters is performed.
2. The system c-based GPU-oriented parameter management method of claim 1, wherein: the step 1 comprises the following steps:
step 11: setting graphfurbar to 1 when graphDrawBar is 0, graphfurnbar is 0, graphdrawfifostatus is non-empty, and graphfurfifostatus is empty; otherwise, executing the step 12;
step 12: setting graphDrawBar to 1 when graphDrawBar is 0, graphfurnbar is 0, graphdrawfafostatus is empty, and graphfurffafostatus is non-empty; otherwise, executing the step 13;
step 13: setting the graphdraw bar to 1 and the graphfunbar to 0 when the graphdraw bar is 0, the graphfunbar is 1, the graphdraw fifo status is empty and the graphfunfifo status is not empty; otherwise, executing the step 14;
step 14: when the graphDrawBar is 1, the graphfurnbar is 0, the graphdrawfafofstatus is non-empty and the graphfurafofstatus is empty, setting the graphDrawBar to 0 and the graphfurnbar to 1; otherwise, executing the step 15;
step 15: reading graphDrawLock, graphFunLock, graphRegLock through the transaction level interface, executing step 16 when the graphRegLock is 0, and executing step 17 when the graphdraglock is 0 and the graphFunLock is 0;
step 16: outputting all variable information and FIFO data information to a graphic management module (2) through a transaction level interface, and executing graphic parameter management;
step 17: all variable information and register window data information are output to a register window management module (3) through a transaction level interface, and register parameter management is executed;
the variable used to represent the barrier state of the graphics drawing command, denoted as graphdragbus, is initialized to 0;
the variable used to represent the status of the graphics function command barrier, noted graphFunBar, is initialized to 0;
the variable used to represent the status of the graphics drawing command FIFO is denoted as graphDrawFifoStatus;
the variable used to represent the status of the graphics function command FIFO is denoted as graphFunFifoStatus.
3. The system c-based GPU-oriented parameter management method of claim 1, wherein: the step 2 comprises the following steps:
step 21: the graph detection module (21) reads a variable graph reglock from the barrier/lock management module (1), and if the variable graph reglock is a lock, the step 21 is circulated; otherwise, executing step 22;
step 22: the graph detection module (21) reads the variable graphDrawBar, graphFunBar, graphDrawFifoStatus, graphFunFifoStatus from the barrier/lock management module (1), and when the graph dragwbar is 0, the graph funnbar is 1, and the graph dragwfifostatus is non-empty, the graph dragwfifostatus is connected to the drawing command processing module (22) through the transaction-level interface, and step 23 is executed; otherwise, when the graphdraw bar is 1, the graphfunbar is 0, and the graphFunFifoStatus is non-empty, connecting to a function code processing module (23) through a transaction-level interface, and executing step 24; otherwise, returning to the step 21;
step 23: the drawing command processing module (22) firstly sets the graphdragwlock to 1, then decodes and assembles the drawing parameters of the graph, and sends the assembled data to the task scheduling module (5) through the object-level interface, and sets the graphdragwlock to 0 after the execution is finished;
step 24: the functional code processing module (23) firstly sets the graphFunLock to 1, then decodes and assembles graphic functional parameters, and transmits the assembled data to the corresponding drawing module (6) through the object-level interface, and sets the graphFunLock to 0 after execution;
the variable used to represent the barrier state of the graphics drawing command, denoted as graphdragbus, is initialized to 0;
the variable used to represent the status of the graphics function command barrier, noted graphFunBar, is initialized to 0;
the variable used to represent the status of the graphics drawing command FIFO is denoted as graphDrawFifoStatus;
the variable used to represent the status of the graphics function command FIFO is denoted as graphFunFifoStatus.
4. The system c-based GPU-oriented parameter management method of claim 1, wherein: the step 3 comprises the following steps:
step 31: setting graphRegLock to 1;
step 32: judging the register read-write type DataType, if the register read-write type DataType is read, executing the step 33; if the write is made, go to step 34;
step 33: connected to an external register module (7) through a transaction-level interface, performing writing of corresponding register data DataStruct according to a register address DataAddr;
step 34: connected to an external register module (7) through a transaction-level interface, executing reading out the corresponding register data DataStruct according to the register address DataAddr;
step 35: the graphRegLock is set to 0.
5. The system c-based GPU-oriented parameter management method of claim 1, wherein: the method comprises the following steps:
two FIFOs are contained inside, receiving two types of commands from the command processing module (4): wherein the drawing commands are stored in a graphics drawing command FIFO and the graphics function commands are stored in a graphics function command FIFO;
the data information is contained in the data information: the register window data specifically comprises a register read-write type DataType, register data DataStruct and a register address DataAddr.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811520556.8A CN109840878B (en) | 2018-12-12 | 2018-12-12 | GPU parameter management method based on SystemC |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811520556.8A CN109840878B (en) | 2018-12-12 | 2018-12-12 | GPU parameter management method based on SystemC |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109840878A CN109840878A (en) | 2019-06-04 |
| CN109840878B true CN109840878B (en) | 2023-05-23 |
Family
ID=66883185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811520556.8A Active CN109840878B (en) | 2018-12-12 | 2018-12-12 | GPU parameter management method based on SystemC |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109840878B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111047504B (en) * | 2019-11-21 | 2023-06-13 | 中国航空工业集团公司西安航空计算技术研究所 | TLM microstructure for GPU sub-image processing based on SystemC |
| CN111880916A (en) * | 2020-07-27 | 2020-11-03 | 长沙景嘉微电子股份有限公司 | Multi-drawing task processing method, device, terminal, medium and host in GPU |
| CN112559139B (en) * | 2020-12-05 | 2022-12-13 | 西安翔腾微电子科技有限公司 | SystemC-based multi-GPU transaction-level model device and operation method |
| CN112581581B (en) * | 2020-12-24 | 2024-04-26 | 西安翔腾微电子科技有限公司 | GPU window transformation module TLM device based on SysML view and operation method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201704528D0 (en) * | 2016-04-07 | 2017-05-03 | Imagination Tech Ltd | Processors supporting atomic writes to multiword memory locations & methods |
| CN106648547A (en) * | 2016-12-12 | 2017-05-10 | 中国航空工业集团公司西安航空计算技术研究所 | Distributed unified management method for GPU graphic state parameters |
| US9876729B1 (en) * | 2015-06-24 | 2018-01-23 | Cadence Design Systems, Inc. | Method and system for efficient data streaming in an emulation system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10275386B2 (en) * | 2014-06-27 | 2019-04-30 | Advanced Micro Devices, Inc. | Memory physical layer interface logic for generating dynamic random access memory (DRAM) commands with programmable delays |
| US10425275B2 (en) * | 2015-02-12 | 2019-09-24 | Advanced Micro Devices, Inc. | Centralized distribution of configuration parameters for a cluster server |
-
2018
- 2018-12-12 CN CN201811520556.8A patent/CN109840878B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9876729B1 (en) * | 2015-06-24 | 2018-01-23 | Cadence Design Systems, Inc. | Method and system for efficient data streaming in an emulation system |
| GB201704528D0 (en) * | 2016-04-07 | 2017-05-03 | Imagination Tech Ltd | Processors supporting atomic writes to multiword memory locations & methods |
| CN106648547A (en) * | 2016-12-12 | 2017-05-10 | 中国航空工业集团公司西安航空计算技术研究所 | Distributed unified management method for GPU graphic state parameters |
Non-Patent Citations (1)
| Title |
|---|
| 嵌入式图形处理器设计;阙恒;《中国优秀硕士学位论文全文数据库信息科技辑》;20080115(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109840878A (en) | 2019-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840878B (en) | GPU parameter management method based on SystemC | |
| US7788625B1 (en) | Method and apparatus for precharacterizing systems for use in system level design of integrated circuits | |
| JP4994393B2 (en) | System and method for generating multiple models at different levels of abstraction from a single master model | |
| CN101872369B (en) | Adaptive state-to-symbolic transformation in a canonical representation | |
| US20070156378A1 (en) | System and method for verification aware synthesis | |
| US20230342283A1 (en) | Waveform based reconstruction for emulation | |
| WO2020083050A1 (en) | Data stream processing method and related device | |
| CN117237513A (en) | Ray tracing system and method, and method for processing data in ray tracing system | |
| US9449131B2 (en) | Extracting system architecture in high level synthesis | |
| CN115618801B (en) | Cache consistency checking method and device and electronic equipment | |
| KR20220148913A (en) | Machine Learning Based Metric Prediction in Early Stage Circuit Design | |
| US7451070B2 (en) | Optimal bus operation performance in a logic simulation environment | |
| CN109727186B (en) | SystemC-based GPU (graphics processing Unit) fragment coloring task scheduling method | |
| CN102184290A (en) | Cycle accurate and bit accurate system level model for embedded micro-processor | |
| CN105447213B (en) | Method and apparatus for being emulated to circuit design | |
| US10162917B1 (en) | Method and system for implementing selective transformation for low power verification | |
| US20200371843A1 (en) | Framework for application driven exploration and optimization of hardware engines | |
| CN111028132B (en) | System C-based GPU command processor unit hardware TLM microstructure | |
| US10380295B1 (en) | Methods, systems, and articles of manufacture for X-behavior verification of an electronic design | |
| US7130784B2 (en) | Logic simulation | |
| WO2021068253A1 (en) | Customized data stream hardware simulation method and apparatus, device, and storage medium | |
| US8819612B1 (en) | Analyzing timing requirements of a hierarchical integrated circuit design | |
| Beyer | Rabbit: Verification of Real Time Systems | |
| US20230342538A1 (en) | Model-driven approach for failure mode, effects, and diagnostic analysis (fmeda) automation for hardware intellectual property of complex electronic systems | |
| CN112559139B (en) | SystemC-based multi-GPU transaction-level model device and operation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |