CN111324572A - Chip adopting heterogeneous system - Google Patents
Chip adopting heterogeneous system Download PDFInfo
- Publication number
- CN111324572A CN111324572A CN201811535913.8A CN201811535913A CN111324572A CN 111324572 A CN111324572 A CN 111324572A CN 201811535913 A CN201811535913 A CN 201811535913A CN 111324572 A CN111324572 A CN 111324572A
- Authority
- CN
- China
- Prior art keywords
- chip
- protocol
- input
- output interface
- dedicated
- 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.)
- Pending
Links
- 101100382264 Mus musculus Ca14 gene Proteins 0.000 claims abstract description 57
- 101100112373 Mus musculus Ctsm gene Proteins 0.000 claims abstract description 57
- 101100094962 Salmo salar salarin gene Proteins 0.000 claims abstract description 57
- 230000015654 memory Effects 0.000 claims abstract description 55
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 15
- 230000002093 peripheral effect Effects 0.000 claims description 7
- 238000003860 storage Methods 0.000 description 12
- 238000013461 design Methods 0.000 description 11
- 230000008901 benefit Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 108091006146 Channels Proteins 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000004806 packaging method and process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 102100021624 Acid-sensing ion channel 1 Human genes 0.000 description 1
- 101710099904 Acid-sensing ion channel 1 Proteins 0.000 description 1
- 102100022094 Acid-sensing ion channel 2 Human genes 0.000 description 1
- 101710099902 Acid-sensing ion channel 2 Proteins 0.000 description 1
- 102100022097 Acid-sensing ion channel 3 Human genes 0.000 description 1
- 101710099898 Acid-sensing ion channel 3 Proteins 0.000 description 1
- 101100217231 Caenorhabditis elegans asic-1 gene Proteins 0.000 description 1
- 101100163897 Caenorhabditis elegans asic-2 gene Proteins 0.000 description 1
- 101100494773 Caenorhabditis elegans ctl-2 gene Proteins 0.000 description 1
- 101100112369 Fasciola hepatica Cat-1 gene Proteins 0.000 description 1
- 101100005271 Neurospora crassa (strain ATCC 24698 / 74-OR23-1A / CBS 708.71 / DSM 1257 / FGSC 987) cat-1 gene Proteins 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007723 transport mechanism Effects 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
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7807—System on chip, i.e. computer system on a single chip; System in package, i.e. computer system on one or more chips in a single package
- G06F15/781—On-chip cache; Off-chip memory
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Communication Control (AREA)
Abstract
本申请公开了一种采用异构系统的芯片。所述芯片包括:CPU;第一输入/输出接口,与窄带物联网NB‑IoT协议的专用芯片相连;第二输入/输出接口,与等级CATM协议的专用芯片相连;数据信道处理DSP芯片,与所述第一输入/输出接口和所述第二输入/输出接口以及CPU相连,根据CPU发送的控制指令,对从所述第一输入/输出接口或所述第二输入/输出接口的信道进行处理;其中,所述DSP芯片中的数据紧耦合存储器、指令紧耦合存储器和一级缓存器中的至少一个的容量大小是根据CATM协议对DSP的硬件需求进行配置的。
The present application discloses a chip using a heterogeneous system. The chip includes: a CPU; a first input/output interface, which is connected to a special chip of the narrowband Internet of Things NB-IoT protocol; a second input/output interface, which is connected to a special chip of the grade CATM protocol; data channel processing DSP chip, and The first input/output interface is connected to the second input/output interface and the CPU, and according to the control instruction sent by the CPU, the channel from the first input/output interface or the second input/output interface is performed. processing; wherein, the capacity of at least one of the data tightly coupled memory, the instruction tightly coupled memory and the first level buffer in the DSP chip is configured according to the hardware requirements of the DSP according to the CATM protocol.
Description
技术领域technical field
本申请涉及信息处理领域,尤指一种采用异构系统的芯片。The present application relates to the field of information processing, in particular to a chip using a heterogeneous system.
背景技术Background technique
随着通信和网络技术的发展,使用新的工艺技术和集成多核处理器方法已经被广泛用来提升处理器的计算性能。然而,随着制造工艺逐渐逼近物理极限,每颗芯片的开发成本逐渐增加,传统依靠不断压榨工艺和集成多核处理器来获取计算性能的效果逐渐减弱,在这种情况下,人们开始希望通过改变设计,注重优化效率来尽可能的获取目前处理器架构的最大性能。With the development of communication and network technology, the use of new process technologies and methods of integrating multi-core processors have been widely used to enhance the computing performance of processors. However, as the manufacturing process gradually approaches the physical limit, the development cost of each chip gradually increases, and the traditional reliance on continuous squeezing processes and integrated multi-core processors to obtain computing performance gradually weakens. In this case, people began to hope that by changing Design, focusing on optimizing efficiency to obtain the maximum performance of the current processor architecture as much as possible.
最原始的SOC架构是使用单个CPU或CPU集群来处理所有的业务,这是同构计算,是使用相同类型的指令集和体系架构处理所有任务。但是同构计算存在局限性。CPU的微架构决定了CPU的专长是处理分支预测、乱序执行以及存储访问等控制类业务。由于CPU内部计算单元很少,对大量数据并行计算,乘累加计算等,只能通过提高CPU的运行频率或者增加CPU的数量来实现。但这两种方法对工艺的要求很高,因此会同时导致CPU的功耗和成本增加。为了解决此问题,引入了异构计算,即用不同类型的指令集和体系架构的计算单元,实现并行处理的方式。异构计算系统有如下几种:The most primitive SOC architecture is to use a single CPU or CPU cluster to process all business. This is homogeneous computing, which uses the same type of instruction set and architecture to process all tasks. But isomorphic computing has limitations. The micro-architecture of the CPU determines that the CPU's expertise is to handle control services such as branch prediction, out-of-order execution, and storage access. Since there are few computing units in the CPU, parallel computing of a large amount of data, multiply-accumulate computing, etc., can only be achieved by increasing the operating frequency of the CPU or increasing the number of CPUs. However, these two methods have high requirements on the process, so they will increase the power consumption and cost of the CPU at the same time. In order to solve this problem, heterogeneous computing is introduced, that is, a method of implementing parallel processing with computing units of different types of instruction sets and architectures. Heterogeneous computing systems are as follows:
1.CPU+GPU的异构系统。这种异构系统充分利用CPU的控制业务能力和GPU的并行处理能力来提升整个系统的计算性能。典型的AMD就是用CPU+GPU的异构系统。但是,因为GPU的算法关联性较少,若所使用的算法不是串行的,GPU就不能很好的发挥并行计算能力。1. Heterogeneous system of CPU+GPU. This heterogeneous system makes full use of the control business capability of the CPU and the parallel processing capability of the GPU to improve the computing performance of the entire system. A typical AMD is a heterogeneous system with CPU+GPU. However, because the algorithms of the GPU are less related, if the algorithms used are not serial, the GPU cannot exert the parallel computing power well.
2.CPU+FPGA的异构系统。FPGA不仅支持数据并行,还支持流水线并行,这种并行计算解决了GPU存在的问题。Intel典型的CPU内嵌FPGA就实现了这种异构系统架构。但同时,单颗FPGA的价格昂贵,量产的话成本会增加,对于市场上的小公司,使用CPU+FPGA的芯片可行性不大。2. Heterogeneous system of CPU+FPGA. FPGA not only supports data parallelism, but also supports pipeline parallelism, which solves the problems of GPU. Intel's typical CPU embedded FPGA implements this heterogeneous system architecture. But at the same time, the price of a single FPGA is expensive, and the cost of mass production will increase. For small companies in the market, it is not feasible to use CPU+FPGA chips.
为了平衡以上两种典型架构的缺点,ASIC将会是异构计算的趋势。In order to balance the shortcomings of the above two typical architectures, ASIC will be the trend of heterogeneous computing.
物联网(Internet of Things,IoT),最初在1999年被提及,即通过射频识别(RFID,RFID+互联网)、红外感应器、全球定位系统,激光扫描器,气体感应器等信息传感设备,按约定的协议,把任何物品与互联网连接起来,进行信息交换和通讯,以实现智能化识别,定位,跟踪,监控和管理的一种网络。简单来说,物联网就是物物相连的互联网。The Internet of Things (IoT) was first mentioned in 1999, that is, through radio frequency identification (RFID, RFID + Internet), infrared sensors, global positioning systems, laser scanners, gas sensors and other information sensing devices, According to the agreed protocol, any item is connected to the Internet for information exchange and communication, so as to realize intelligent identification, positioning, tracking, monitoring and management of a network. Simply put, the Internet of Things is the Internet of things connected.
3GPPR13制定的CAT-M1(category,等级)和NB-IoT(Narrow BandInternetofThings,窄带物联网)标准是目前物联网技术里面最火热的两个标准。CAT-M1和NB-IoT都被认为是最具潜力的物联网技术,这两种技术适用于不同的应用场景,但很多人从覆盖范围,成本和功耗方面,普遍认为NB-IoT比CAT-M1更有优势,目前芯片市场出现的物联网芯片也都基于NB-IoT标准。但是有资料显示,CAT-M1在覆盖范围和功耗上实际要比NB-IoT更优,并且CATM-M1的速率更高。因此,基于CAT-M1标准的物联网芯片在不久就会出现。The CAT-M1 (category, level) and NB-IoT (Narrow Band Internet of Things, Narrow Band Internet of Things) standards formulated by 3GPPR13 are the two most popular standards in the current Internet of Things technology. Both CAT-M1 and NB-IoT are considered to be the most potential IoT technologies. These two technologies are suitable for different application scenarios, but many people generally believe that NB-IoT is better than CAT in terms of coverage, cost and power consumption. -M1 has more advantages, and the IoT chips currently appearing in the chip market are also based on the NB-IoT standard. However, some data show that CAT-M1 is actually better than NB-IoT in terms of coverage and power consumption, and the rate of CATM-M1 is higher. Therefore, IoT chips based on the CAT-M1 standard will appear in the near future.
目前我们已经成功完成的物联网芯片是基于NB-IoT标准,并且正在预研和评估CATM的物联网芯片。在物联网时代,随着数字化转型,它需要更敏捷地连接,也需要更有效地数据处理。At present, the IoT chip we have successfully completed is based on the NB-IoT standard, and we are pre-researching and evaluating the IoT chip of CATM. In the age of IoT, with digital transformation, it needs to be more agile to connect and also to process data more efficiently.
发明内容SUMMARY OF THE INVENTION
为了解决上述技术问题,本申请提供了一种采用异构系统的芯片,能够在保证性能的前提下,降低芯片的硬件成本。In order to solve the above technical problems, the present application provides a chip using a heterogeneous system, which can reduce the hardware cost of the chip on the premise of ensuring performance.
为了达到本申请目的,本申请提供了一种采用异构系统的芯片,包括:In order to achieve the purpose of the application, the application provides a chip using a heterogeneous system, including:
CPU;CPU;
第一输入/输出接口,与NB-IoT协议的专用芯片相连;The first input/output interface is connected to the dedicated chip of the NB-IoT protocol;
第二输入/输出接口,与CATM协议的专用芯片相连;The second input/output interface is connected with the dedicated chip of the CATM protocol;
数据信道处理DSP芯片,与所述第一输入/输出接口和所述第二输入/输出接口以及CPU相连,根据CPU发送的控制指令,对从所述第一输入/输出接口或所述第二输入/输出接口的信道进行处理;The data channel processing DSP chip is connected with the first input/output interface and the second input/output interface and the CPU, and according to the control instruction sent by the CPU, the data from the first input/output interface or the second input/output interface is The channel of the input/output interface is processed;
其中,所述DSP芯片中的数据紧耦合存储器、指令紧耦合存储器和一级缓存器中的至少一个的容量大小是根据CATM协议对DSP的硬件需求进行配置的。Wherein, the capacity of at least one of the data tightly coupled memory, the instruction tightly coupled memory and the first level buffer in the DSP chip is configured according to the hardware requirements of the DSP according to the CATM protocol.
在一个示例性实施例中,所述数据紧耦合存储器和/或所述指令紧耦合存储器的个数为至少两个;In an exemplary embodiment, the number of the data tightly coupled memory and/or the instruction tightly coupled memory is at least two;
其中,在运行NB-IoT协议的专用芯片时,所述数据紧耦合存储器和/或所述指令紧耦合存储器中的一部分处于工作状态,剩余部分处于关闭状态;在运行CATM协议的专用芯片时,所述数据紧耦合存储器和/或所述指令紧耦合存储器中的全部均处于工作状态。Wherein, when running the dedicated chip of the NB-IoT protocol, a part of the data tightly coupled memory and/or the instruction tightly coupled memory is in a working state, and the rest is in a closed state; when running the dedicated chip of the CATM protocol, All of the data tightly coupled memory and/or the instruction tightly coupled memory are in a working state.
在一个示例性实施例中,在运行NB-IoT协议的专用芯片时,所述一级缓存器处于关闭状态;所述第二缓存器处于工作状态。In an exemplary embodiment, when a dedicated chip of the NB-IoT protocol is running, the first level buffer is in a closed state; the second buffer is in a working state.
在一个示例性实施例中,所述DSP的信号频率包括第一频率和第二频率;其中所述第一频率在预先设置的低频范围内,所述第二频率在预先设置的高频范围内;In an exemplary embodiment, the signal frequency of the DSP includes a first frequency and a second frequency; wherein the first frequency is within a preset low frequency range, and the second frequency is within a preset high frequency range ;
在运行NB-IoT协议的专用芯片时,所述DSP采用第一频率对所述第一输入/输出接口接收的数据处理;在运行CATM协议的专用芯片时,所述DSP采用第二频率对所述第二输入/输出接口输入的数据进行处理。When running the dedicated chip of the NB-IoT protocol, the DSP uses the first frequency to process the data received by the first input/output interface; when running the dedicated chip of the CATM protocol, the DSP uses the second frequency to process the data received by the first input/output interface. The data input from the second input/output interface is processed.
在一个示例性实施例中,所述芯片还包括:In an exemplary embodiment, the chip further includes:
控制器,控制所述NB-IoT协议的专用芯片和CATM协议的专用芯片中的一个处于工作状态。The controller controls one of the dedicated chip of the NB-IoT protocol and the dedicated chip of the CATM protocol to be in a working state.
在一个示例性实施例中,所述CPU与所述DSP与外设的系统级芯片构成第一芯片,并通过预留的第一输入/输出接口与NB-IoT协议的专用芯片相连,通过预留的第二输入/输出接口与CATM协议的专用芯片相连。In an exemplary embodiment, the CPU, the DSP and the system-on-chip of peripherals constitute a first chip, and are connected to the dedicated chip of the NB-IoT protocol through a reserved first input/output interface, The reserved second input/output interface is connected with the dedicated chip of the CATM protocol.
在一个示例性实施例中,所述第一芯片还包括:In an exemplary embodiment, the first chip further includes:
存储器,存储NB-IoT协议的专用芯片和/或CATM协议的专用芯片的配置信息;Memory, which stores the configuration information of the dedicated chip for the NB-IoT protocol and/or the dedicated chip for the CATM protocol;
其中,所述CPU利用所述存储器内的配置信息对NB-IoT协议的专用芯片和/或CATM协议的专用芯片进行配置。Wherein, the CPU configures the dedicated chip of the NB-IoT protocol and/or the dedicated chip of the CATM protocol by using the configuration information in the memory.
在一个示例性实施例中,所述第一芯片、所述NB-IoT协议的专用芯片、CATM协议的专用芯片合封在一起。In an exemplary embodiment, the first chip, the dedicated chip for the NB-IoT protocol, and the dedicated chip for the CATM protocol are packaged together.
在一个示例性实施例中,所述第一芯片还包括:In an exemplary embodiment, the first chip further includes:
一个或至少两个输入/输出接口,其中每个输入/输出接口与一个标准或算法对应的专用芯片相连;One or at least two input/output interfaces, wherein each input/output interface is connected to a dedicated chip corresponding to a standard or algorithm;
其中,所述CPU控制所述标准或算法对应的专用芯片与对应的输入/输出接口的连接状态。Wherein, the CPU controls the connection state between the dedicated chip corresponding to the standard or the algorithm and the corresponding input/output interface.
在一个示例性实施例中,所述第一芯片还包括:In an exemplary embodiment, the first chip further includes:
存储器,存储所述标准或算法对应的专用芯片的配置信息;a memory, storing configuration information of the dedicated chip corresponding to the standard or algorithm;
其中,所述CPU利用所述存储器内的配置信息对所述标准或算法对应的专用芯片进行配置。The CPU configures the dedicated chip corresponding to the standard or algorithm by using the configuration information in the memory.
本申请提供的实施例,通过两个输入接口连接NB-IoT协议的专用芯片和CATM协议的专用芯片,实现在一个芯片上实现两个协议所需功能的目的,实现两个专用芯片使用同一组CPU和DSP的目的,节省了硬件成本,另外,通过对DSP中硬件的容量的调整,有效保证了CATM协议运行时的性能,实现在保证芯片的硬件性能的前提下,降低了芯片的硬件成本。In the embodiment provided in this application, the special chip of the NB-IoT protocol and the special chip of the CATM protocol are connected through two input interfaces, so as to realize the purpose of realizing the functions required by the two protocols on one chip, and realize that the two special chips use the same group The purpose of CPU and DSP saves the hardware cost. In addition, by adjusting the hardware capacity in the DSP, the performance of the CATM protocol during operation is effectively guaranteed, and the hardware cost of the chip is reduced on the premise of ensuring the hardware performance of the chip. .
本申请的其它特征和优点将在随后的说明书中阐述,并且,部分地从说明书中变得显而易见,或者通过实施本申请而了解。本申请的目的和其他优点可通过在说明书、权利要求书以及附图中所特别指出的结构来实现和获得。Other features and advantages of the present application will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the present application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the description, claims and drawings.
附图说明Description of drawings
附图用来提供对本申请技术方案的进一步理解,并且构成说明书的一部分,与本申请的实施例一起用于解释本申请的技术方案,并不构成对本申请技术方案的限制。The accompanying drawings are used to provide a further understanding of the technical solutions of the present application, and constitute a part of the specification. They are used to explain the technical solutions of the present application together with the embodiments of the present application, and do not constitute a limitation on the technical solutions of the present application.
图1为本发明提供的采用异构系统的芯片的结构图;1 is a structural diagram of a chip using a heterogeneous system provided by the present invention;
图2为相关技术中基于NB-IoT标准的物联网异构系统的示意图;2 is a schematic diagram of a heterogeneous system of the Internet of Things based on the NB-IoT standard in the related art;
图3为本申请提供的CATM标准的物联网异构系统的示意图;3 is a schematic diagram of a heterogeneous system of the Internet of Things of the CATM standard provided by the present application;
图4为本申请提供的CPU+DSP+ASIC异构物联网系统的示意图;4 is a schematic diagram of a CPU+DSP+ASIC heterogeneous IoT system provided by the present application;
图5为本申请提供的CPU+DSP+ASIC异构系统的示意图。FIG. 5 is a schematic diagram of a CPU+DSP+ASIC heterogeneous system provided by the present application.
具体实施方式Detailed ways
为使本申请的目的、技术方案和优点更加清楚明白,下文中将结合附图对本申请的实施例进行详细说明。需要说明的是,在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互任意组合。In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be noted that, the embodiments in the present application and the features in the embodiments may be arbitrarily combined with each other if there is no conflict.
在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行。并且,虽然在流程图中示出了逻辑顺序,但是在某些情况下,可以以不同于此处的顺序执行所示出或描述的步骤。The steps shown in the flowcharts of the figures may be performed in a computer system, such as a set of computer-executable instructions. Also, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that herein.
图1为本发明提供的采用异构系统的芯片的结构图。图1所示芯片包括:FIG. 1 is a structural diagram of a chip using a heterogeneous system provided by the present invention. The chip shown in Figure 1 includes:
CPU;CPU;
第一输入/输出接口,与NB-IoT协议的专用芯片相连;The first input/output interface is connected to the dedicated chip of the NB-IoT protocol;
第二输入/输出接口,与CATM协议的专用芯片相连;The second input/output interface is connected with the dedicated chip of the CATM protocol;
数据信道处理DSP芯片,与所述第一输入/输出接口和所述第二输入/输出接口以及CPU相连,根据CPU发送的控制指令,对从所述第一输入/输出接口或所述第二输入/输出接口的信道进行处理;The data channel processing DSP chip is connected with the first input/output interface and the second input/output interface and the CPU, and according to the control instruction sent by the CPU, the data from the first input/output interface or the second input/output interface is The channel of the input/output interface is processed;
其中,所述DSP芯片中的数据紧耦合存储器、指令紧耦合存储器和一级缓存器中的至少一个的容量大小是根据CATM协议对DSP的硬件需求进行配置的。Wherein, the capacity of at least one of the data tightly coupled memory, the instruction tightly coupled memory and the first level buffer in the DSP chip is configured according to the hardware requirements of the DSP according to the CATM protocol.
在本示例性实施例中,通过两个输入接口连接NB-IoT协议的专用芯片和CATM协议的专用芯片,实现在一个芯片上实现两个协议所需功能的目的;在确认芯片能够执行NB-IoT和CATM两种协议时,获取上述两个协议对物理层DSP的需求信息,选择对DSP性能高的需求信息进行硬件配置。比如,CATM峰值速率是375Kbps,NB的峰值速率大约是50Kbps。因此CATM对DSP的性能要求更高。从DSP的硬件配置上,一方面可以增大紧耦合存储器(TightlyCoupled Memories,TCM)的容量,这样可以充分利用TCM的优势,减小访问存储的延时来提高整体性能;另一方面可以增加一级缓存的容量,这样动态访问外部存储器可以提高取指性能的基础上减小片上存储的面积。这两方面存储器的增加大小依据软件的需求及性能来平衡。In this exemplary embodiment, the dedicated chip of the NB-IoT protocol and the dedicated chip of the CATM protocol are connected through two input interfaces, so as to realize the purpose of realizing the functions required by the two protocols on one chip; after confirming that the chip can execute the NB-IoT protocol When there are two protocols, IoT and CATM, obtain the demand information of the above two protocols on the physical layer DSP, and select the demand information with high DSP performance for hardware configuration. For example, the peak rate of CATM is 375Kbps, and the peak rate of NB is about 50Kbps. Therefore, CATM has higher requirements on the performance of DSP. From the hardware configuration of DSP, on the one hand, the capacity of Tightly Coupled Memories (TCM) can be increased, so that the advantages of TCM can be fully utilized, and the delay of accessing memory can be reduced to improve the overall performance; on the other hand, a Level cache capacity, so that dynamic access to external memory can improve instruction fetch performance and reduce the area of on-chip storage. The increased size of these two aspects is balanced by software requirements and performance.
增加的DSP的硬件性能的成本要低于增设一个CPU和一个DSP的成本,因此,在保证硬件性能的前提下,降低了硬件成本。The cost of increasing the hardware performance of the DSP is lower than the cost of adding a CPU and a DSP. Therefore, the hardware cost is reduced on the premise of ensuring the hardware performance.
本申请提供的芯片实施例,通过两个输入接口连接NB-IoT协议的专用芯片和CATM协议的专用芯片,实现在一个芯片上实现两个协议所需功能的目的,实现两个专用芯片使用同一组CPU和DSP的目的,节省了硬件成本,另外,通过对DSP中硬件的容量的调整,有效保证了CATM协议运行时的性能,实现在保证芯片的硬件性能的前提下,降低了芯片的硬件成本。In the chip embodiment provided in this application, the special chip of the NB-IoT protocol and the special chip of the CATM protocol are connected through two input interfaces, so as to realize the purpose of realizing the functions required by the two protocols on one chip, and realize that the two special chips use the same chip. The purpose of combining CPU and DSP saves the hardware cost. In addition, by adjusting the hardware capacity in the DSP, the performance of the CATM protocol during operation is effectively guaranteed, and the hardware performance of the chip is guaranteed, and the hardware performance of the chip is reduced. cost.
下面对本申请提供的技术特征进行说明:The technical features provided by the application are described below:
在一个示例性实施例中,所述数据紧耦合存储器和/或所述指令紧耦合存储器的个数为至少两个;In an exemplary embodiment, the number of the data tightly coupled memory and/or the instruction tightly coupled memory is at least two;
其中,在运行NB-IoT协议的专用芯片时,所述数据紧耦合存储器和/或所述指令紧耦合存储器中的一部分处于工作状态,剩余部分处于关闭状态;在运行CATM协议的专用芯片时,所述数据紧耦合存储器和/或所述指令紧耦合存储器中的全部均处于工作状态。Wherein, when running the dedicated chip of the NB-IoT protocol, a part of the data tightly coupled memory and/or the instruction tightly coupled memory is in a working state, and the rest is in a closed state; when running the dedicated chip of the CATM protocol, All of the data tightly coupled memory and/or the instruction tightly coupled memory are in a working state.
在本示例性实施例中,所述数据紧耦合存储器和/或所述指令紧耦合存储器的个数为至少两个,目的在于保证在运行CATM协议的专用芯片时,本申请提供芯片的硬件性能,而在运行NB-IoT协议的专用芯片时,并不存在性能不够的问题,因此,所述数据紧耦合存储器和/或所述指令紧耦合存储器中的一部分处于工作状态,剩余部分处于关闭状态,减少芯片的功耗。In this exemplary embodiment, the number of the data tightly coupled memory and/or the instruction tightly coupled memory is at least two, the purpose is to ensure that when the dedicated chip of the CATM protocol is running, the hardware performance of the chip is provided by the present application , and when running the dedicated chip of the NB-IoT protocol, there is no problem of insufficient performance. Therefore, a part of the data tightly coupled memory and/or the instruction tightly coupled memory is in a working state, and the rest is in a closed state. , reduce the power consumption of the chip.
在一个示例性实施例中,在运行NB-IoT协议的专用芯片时,所述一级缓存器处于关闭状态;所述第二缓存器处于工作状态。In an exemplary embodiment, when a dedicated chip of the NB-IoT protocol is running, the first level buffer is in a closed state; the second buffer is in a working state.
在本示例性实施例中,一级缓存器的设置是为了保证在运行CATM协议的专用芯片时,本申请提供芯片的硬件性能,而在运行NB-IoT协议的专用芯片时,是不需要一级缓存器,可以关闭该一级缓存器,减少芯片的功耗。In this exemplary embodiment, the setting of the first-level buffer is to ensure that the application provides the hardware performance of the chip when running the dedicated chip of the CATM protocol, but does not need a dedicated chip when running the dedicated chip of the NB-IoT protocol. The first-level buffer can be closed to reduce the power consumption of the chip.
在一个示例性实施例中,所述DSP的信号频率包括第一频率和第二频率;其中所述第一频率在预先设置的低频范围内,所述第二频率在预先设置的高频范围内;In an exemplary embodiment, the signal frequency of the DSP includes a first frequency and a second frequency; wherein the first frequency is within a preset low frequency range, and the second frequency is within a preset high frequency range ;
在运行NB-IoT协议的专用芯片时,所述DSP采用第一频率对所述第一输入/输出接口接收的数据处理;在运行CATM协议的专用芯片时,所述DSP采用第二频率对所述第二输入/输出接口输入的数据进行处理。When running the dedicated chip of the NB-IoT protocol, the DSP uses the first frequency to process the data received by the first input/output interface; when running the dedicated chip of the CATM protocol, the DSP uses the second frequency to process the data received by the first input/output interface. The data input from the second input/output interface is processed.
通过设置DSP的信号频率是可调的,方便上述两个协议在所需的信号频率中工作,保证芯片的工作性能。By setting the signal frequency of the DSP to be adjustable, it is convenient for the above two protocols to work in the required signal frequency, and the working performance of the chip is guaranteed.
在一个示例性实施例中,所述芯片还包括:In an exemplary embodiment, the chip further includes:
控制器,控制所述NB-IoT协议的专用芯片和CATM协议的专用芯片中的一个处于工作状态。The controller controls one of the dedicated chip of the NB-IoT protocol and the dedicated chip of the CATM protocol to be in a working state.
在本示例性例实施例中,所述控制器可以为一个软件模块,集成于CPU上,从而有效控制芯片的硬件成本。In this exemplary embodiment, the controller may be a software module integrated on the CPU, thereby effectively controlling the hardware cost of the chip.
当使用NB-IOT协议的时候,使用dtcm0和itcm0,在这种场景下,dctm1和itcm1处于shutdown关电模式来降功耗。一级缓存icache和dcache是否使用取决于软件需求,如果不使用,对应的存储也做shutdown模式。当使用CATM协议的时候,使用dtcm0+dtcm1,itcm0+itcm1和icache+dcache。When using the NB-IOT protocol, use dtcm0 and itcm0. In this scenario, dctm1 and itcm1 are in shutdown mode to reduce power consumption. Whether the first-level cache icache and dcache are used depends on software requirements. If they are not used, the corresponding storage is also in shutdown mode. When using the CATM protocol, use dtcm0+dtcm1, itcm0+itcm1 and icache+dcache.
在一个示例性实施例中,所述CPU与所述DSP与外设的系统级芯片构成第一芯片,并通过预留的第一输入/输出接口与NB-IoT协议的专用芯片相连,通过预留的第二输入/输出接口与CATM协议的专用芯片相连。In an exemplary embodiment, the CPU, the DSP and the system-on-chip of peripherals constitute a first chip, and are connected to the dedicated chip of the NB-IoT protocol through a reserved first input/output interface, The reserved second input/output interface is connected with the dedicated chip of the CATM protocol.
在一个示例性实施例中,所述第一芯片还包括:In an exemplary embodiment, the first chip further includes:
存储器,存储NB-IoT协议的专用芯片和/或CATM协议的专用芯片的配置信息;Memory, which stores the configuration information of the dedicated chip for the NB-IoT protocol and/or the dedicated chip for the CATM protocol;
其中,所述CPU利用所述存储器内的配置信息对NB-IoT协议的专用芯片和/或CATM协议的专用芯片进行配置。Wherein, the CPU configures the dedicated chip of the NB-IoT protocol and/or the dedicated chip of the CATM protocol by using the configuration information in the memory.
在一个示例性实施例中,所述第一芯片、所述NB-IoT协议的专用芯片、CATM协议的专用芯片合封在一起。In an exemplary embodiment, the first chip, the dedicated chip for the NB-IoT protocol, and the dedicated chip for the CATM protocol are packaged together.
3GPP定义的网络接入技术标准从LTE CAT1到CAT10。标准的演进,应用的场景不同,但只涉及部分硬件逻辑的变动和相关软件的变动。对部分硬件逻辑变动也同样可以做成专用的ASIC芯片,对接到固定芯片中。如图4所示。这种方式同样可以减小对协议演进做优化的芯片成本。The network access technology standards defined by 3GPP range from LTE CAT1 to CAT10. The evolution of the standard has different application scenarios, but only involves some hardware logic changes and related software changes. For some hardware logic changes, it can also be made into a dedicated ASIC chip and connected to the fixed chip. As shown in Figure 4. In this way, the chip cost for optimizing protocol evolution can also be reduced.
在一个示例性实施例中,在芯片设计阶段,将固定芯片与基于NB物联网算法的专用芯片分开设计,并且在两个芯片上都预留IO接口。NB专用芯片只用来实现算法和标准,本身面积并不大,有可重复设计的优势。对NB专用芯片的配置做在固定芯片中。封装的时候两者之间通过IO口做合封。当3GPPR14对NB标准有更新的时候,可以只对NB专用芯片进行重设计,然后重新进行封装。这样可减小芯片优化的设计复杂度和成本。In an exemplary embodiment, in the chip design stage, the fixed chip is designed separately from the dedicated chip based on the NB IoT algorithm, and IO interfaces are reserved on both chips. The NB-specific chip is only used to implement algorithms and standards, and its area is not large, so it has the advantage of repeatable design. The configuration of the NB-specific chip is done in the fixed chip. When packaging, the two are sealed together through the IO port. When 3GPPR14 updates the NB standard, only the NB-specific chip can be redesigned and then repackaged. This reduces the design complexity and cost of chip optimization.
在一个示例性实施例中,在芯片设计阶段,将固定芯片与两个专用芯片分开设计,并且在每个芯片上都预留IO接口。两个专用芯片分别用来实现NB和CATM的算法和标准,本身面积并不大,有可重复设计的优势。对专用芯片的配置做在固定芯片中,封装的时候固定芯片与两个专用芯片之间通过IO口做合封。因为不同的标准或算法实际中的应用场景不同,实际芯片使用过程中,可以灵活的使用CPU,通过软件配置固定芯片中的配置文件来选择NB或者CATM的标准或算法对应的专用芯片。In an exemplary embodiment, in the chip design stage, the fixed chip and the two dedicated chips are designed separately, and IO interfaces are reserved on each chip. The two special-purpose chips are used to implement the algorithms and standards of NB and CATM respectively. The area itself is not large, and it has the advantage of repeatable design. The configuration of the special chip is done in the fixed chip, and the fixed chip and the two special chips are sealed through the IO port during packaging. Because the actual application scenarios of different standards or algorithms are different, the CPU can be used flexibly during the actual use of the chip, and the configuration file in the fixed chip can be configured by software to select the special chip corresponding to the standard or algorithm of NB or CATM.
在一个示例性实施例中,所述第一芯片还包括:In an exemplary embodiment, the first chip further includes:
一个或至少两个输入/输出接口,其中每个输入/输出接口与一个标准或算法对应的专用芯片相连;One or at least two input/output interfaces, wherein each input/output interface is connected to a dedicated chip corresponding to a standard or algorithm;
其中,所述CPU控制所述标准或算法对应的专用芯片与对应的输入/输出接口的连接状态。Wherein, the CPU controls the connection state between the dedicated chip corresponding to the standard or the algorithm and the corresponding input/output interface.
在一个示例性实施例中,所述第一芯片还包括:In an exemplary embodiment, the first chip further includes:
存储器,存储所述标准或算法对应的专用芯片的配置信息;a memory, storing configuration information of the dedicated chip corresponding to the standard or algorithm;
其中,所述CPU利用所述存储器内的配置信息对所述标准或算法对应的专用芯片进行配置。The CPU configures the dedicated chip corresponding to the standard or algorithm by using the configuration information in the memory.
在一个示例性实施例中,在芯片设计阶段,将固定芯片与单个专用芯片分开设计,并且在两个芯片上都预留IO接口。专用芯片只用来实现算法和标准,此标准或算法不局限于物联网标准或算法,本身面积并不大,有可重复设计的优势。对专用芯片的配置做在固定芯片中。封装的时候两者之间通过IO口做合封。当标准或算法有更新的时候,可以只对专用芯片进行重设计,然后重新进行封装。这样可减小芯片优化的设计复杂度和成本。In an exemplary embodiment, in the chip design stage, the fixed chip is designed separately from a single dedicated chip, and IO interfaces are reserved on both chips. Dedicated chips are only used to implement algorithms and standards. This standard or algorithm is not limited to IoT standards or algorithms. The area itself is not large, and it has the advantage of repeatable design. The configuration of the dedicated chip is done in the fixed chip. When packaging, the two are sealed together through the IO port. When standards or algorithms are updated, only the dedicated chip can be redesigned and then repackaged. This reduces the design complexity and cost of chip optimization.
在一个示例性实施例中,在芯片设计阶段,将固定芯片与多个专用芯片分开设计,并且每个芯片上都预留IO接口。多个专用芯片用来实现不同的算法和标准,此标准或算法不局限于物联网标准或算法。本身面积不会太大,有可重复设计的优势。对专用芯片的配置做在固定芯片中,封装的时候固定芯片与专用芯片之前通过IO接口做合封。因为不同的标准或算法实际中的应用场景不同,实际芯片使用过程中,可以灵活的使用CPU,通过软件配置固定芯片中的配置文件来选择不同的标准或算法对应的专用芯片。In an exemplary embodiment, in the chip design stage, the fixed chip is designed separately from a plurality of dedicated chips, and IO interfaces are reserved on each chip. Multiple dedicated chips are used to implement different algorithms and standards, which are not limited to IoT standards or algorithms. The area itself will not be too large, and it has the advantage of repeatable design. The configuration of the special chip is done in the fixed chip, and the fixed chip and the special chip are sealed together through the IO interface before packaging. Because the actual application scenarios of different standards or algorithms are different, the CPU can be flexibly used in the actual chip use process, and the configuration files in the fixed chip can be configured by software to select the dedicated chips corresponding to different standards or algorithms.
下面对本申请提供的硬件结构进行说明:The hardware structure provided by this application is described below:
基于上述提到的物联网芯片和异构计算系统,本发明提出一种基于CPU+DSP+ASIC的异构物联网芯片系统。Based on the above mentioned IoT chip and heterogeneous computing system, the present invention proposes a heterogeneous IoT chip system based on CPU+DSP+ASIC.
图2为相关技术中基于NB-IoT标准的物联网异构系统的示意图;如图2所示,物联网芯片是基于NB-IoT标准,使用CPU+DSP的异构系统。其中,CPU主要处理控制类业务,DSP主要处理协议栈和物理层的运算类业务,总线上挂有IoT硬件模块,实现NB-IoT协议。Figure 2 is a schematic diagram of an IoT heterogeneous system based on the NB-IoT standard in the related art; as shown in Figure 2, the IoT chip is a heterogeneous system based on the NB-IoT standard and using a CPU+DSP. Among them, the CPU mainly handles the control business, the DSP mainly handles the operation business of the protocol stack and the physical layer, and the IoT hardware module is hung on the bus to realize the NB-IoT protocol.
图3为本申请提供的CATM标准的物联网异构系统的示意图。如图3所示,下一代物联网芯片是基于CATM标准制定的,与NB-IoT标准的芯片对比可知,其区别在于:IoT协议硬件的实现不同和物理层软件处理不同以及对物理层处理器DSP的需求不同。FIG. 3 is a schematic diagram of an IoT heterogeneous system of the CATM standard provided by the present application. As shown in Figure 3, the next-generation IoT chip is formulated based on the CATM standard. Compared with the chip of the NB-IoT standard, the difference lies in: the implementation of the IoT protocol hardware is different, the physical layer software processing is different, and the physical layer processor is different. DSP needs are different.
在除去软件部分可编程的因素下,硬件不同点如图3所示的虚线框部分,协议的实现不同体现在IoT硬件模块上和对DSP的需求不同。造成上述不同的原因在于CATM的覆盖范围和速率更高,对处理器运算能力要求更高。在确定原因后,本申请通过增加一级缓存或者增加紧耦合存储器以提升取指效率和运算性能。In addition to the programmable factor of the software part, the hardware difference is shown in the dotted box part in Figure 3. The implementation of the protocol is different in the IoT hardware module and the different requirements for DSP. The reason for the above difference is that the coverage and speed of CATM are higher, and the requirements for processor computing power are higher. After the cause is determined, the present application improves instruction fetch efficiency and computing performance by adding a level-one cache or a tightly coupled memory.
下面以本发明实施例提供的应用实例进行说明:The following describes the application examples provided by the embodiments of the present invention:
图4为本申请提供的CPU+DSP+ASIC异构物联网系统的示意图;如图4所示,各模块的功能如下:Figure 4 is a schematic diagram of a CPU+DSP+ASIC heterogeneous IoT system provided by the application; as shown in Figure 4, the functions of each module are as follows:
CPU:处理应用层控制类业务。CPU: Processes application layer control services.
DSP:处理协议栈和物理层运算类业务,其中dtcm0,dtcm1,itcm0,itcm1分别是数据紧耦合存储器和指令紧耦合存储器。DSP: handles protocol stack and physical layer operation services, where dtcm0, dtcm1, itcm0, and itcm1 are data tightly coupled memory and instruction tightly coupled memory respectively.
Peripheral:外设部分,包括外设IP和外部存储器。Peripheral: Peripheral part, including peripheral IP and external memory.
NB-IoT ASIC:实现NB_IoT标准的专用芯片。NB-IoT ASIC: A dedicated chip that implements the NB_IoT standard.
CATM ASIC:实现CATM标准的专用芯片。CATM ASIC: A dedicated chip that implements the CATM standard.
Config:配置模块,配置使用哪些内部存储器,配置使用哪个专用芯片。Config: Configure the module, configure which internal memory to use, and configure which dedicated chip to use.
在确认执行NB和CATM两种协议时,获取上述两个协议对物理层DSP的需求信息,选择对DSP性能高的需求信息进行硬件配置。比如,CATM峰值速率是375Kbps,NB的峰值速率大约是50Kbps。因此CATM对DSP的性能要求更高。从DSP的硬件配置上,一方面可以增大紧耦合存储器的容量,这样可以充分利用TCM的优势,减小访问存储的延时来提高整体性能;另一方面可以增加一级缓存的容量,这样动态访问外部存储器可以提高取指性能的基础上减小片上存储的面积。这两方面存储器的增加大小依据软件的需求及性能来平衡。When confirming the execution of the NB and CATM protocols, obtain the requirement information of the above two protocols on the physical layer DSP, and select the requirement information with high DSP performance for hardware configuration. For example, the peak rate of CATM is 375Kbps, and the peak rate of NB is about 50Kbps. Therefore, CATM has higher requirements on the performance of DSP. From the hardware configuration of DSP, on the one hand, the capacity of the tightly coupled memory can be increased, so that the advantages of TCM can be fully utilized, and the delay of accessing the memory can be reduced to improve the overall performance; on the other hand, the capacity of the first-level cache can be increased, so that the Dynamic access to external memory can improve instruction fetch performance and reduce on-chip storage area. The increased size of these two aspects is balanced by software requirements and performance.
图5为本申请提供的CPU+DSP+ASIC异构系统的示意图,如图5所示,当使用NB-IOT协议的时候,使用dtcm0和itcm0,在这种场景下,dctm1和itcm1处于shutdown关电模式来降功耗。一级缓存icache和dcache是否使用取决于软件需求,如果不使用,对应的存储也做shutdown模式。当使用CATM协议的时候,使用dtcm0+dtcm1,itcm0+itcm1和icache+dcache。同时,DSP要做频率可调功能,在CATM协议下使用高频,在NB-IOT协议下使用低频。Figure 5 is a schematic diagram of a CPU+DSP+ASIC heterogeneous system provided by this application. As shown in Figure 5, when the NB-IOT protocol is used, dtcm0 and itcm0 are used. In this scenario, dctm1 and itcm1 are in shutdown power mode to reduce power consumption. Whether the first-level cache icache and dcache are used depends on software requirements. If they are not used, the corresponding storage is also in shutdown mode. When using the CATM protocol, use dtcm0+dtcm1, itcm0+itcm1 and icache+dcache. At the same time, the DSP should do the frequency adjustable function, use high frequency under the CATM protocol, and use low frequency under the NB-IOT protocol.
其中,CPU+DSP+外设的SoC架构单独做ASIC 1芯片,预留与专用芯片对接的IO接口,再根据当前的NB-IoT协议开发对应的专用ASIC 2芯片,预留IO接口,以及根据当前的CATM协议开发对应的专用ASIC 3芯片,预留IO接口,在芯片封装的时候,对ASIC1,ASIC2,和ASIC3三个芯片做合封。最后,分别针对CATM和NB的协议开发物理层软件,在芯片做成产品的时候,应用场景具体是使用NB的标准还是CATM的标准,通过Config配置来确定。Among them, the SoC architecture of CPU + DSP + peripherals is used as a separate ASIC 1 chip, and the IO interface for docking with the dedicated chip is reserved, and then the corresponding dedicated ASIC 2 chip is developed according to the current NB-IoT protocol, and the IO interface is reserved. The CATM protocol developed the corresponding dedicated ASIC 3 chip, reserved IO interface, when the chip was packaged, the three chips ASIC1, ASIC2, and ASIC3 were packaged together. Finally, the physical layer software is developed for the CATM and NB protocols respectively. When the chip is made into a product, the application scenario is determined by Config configuration whether to use the NB standard or the CATM standard.
由于物联网标准还在发展过程中,比如NB标准的一些规范会在3GPPR14中制定,而目前实现的都是R13的标准。使用CPU+DSP+ASIC的异构系统,等标准完全确定后,可以只更新物联网协议的ASIC专用芯片以及对应的物理层软件就可以,不用重新做芯片。同时针对不同物联网标准的不同芯片,可以只对物联网标准的硬件实现做成专用芯片,然后根据配置选择不同的标准应用于不同的场景,这样公共的部分可以保持不变,对芯片的下一代优化节省了成本。另外,此异构系统架构,不局限于在物联网芯片中,同时,适用于终端数据卡芯片。As the IoT standard is still in the process of development, for example, some specifications of the NB standard will be formulated in 3GPPR14, and the current implementation is the R13 standard. For heterogeneous systems using CPU + DSP + ASIC, after the standard is completely determined, only the ASIC dedicated chip of the IoT protocol and the corresponding physical layer software can be updated, and there is no need to remake the chip. At the same time, for different chips of different IoT standards, only the hardware implementation of the IoT standard can be made into a dedicated chip, and then different standards can be selected according to the configuration to apply to different scenarios, so that the common part can remain unchanged. One-generation optimization saves costs. In addition, this heterogeneous system architecture is not limited to IoT chips, but is also applicable to terminal data card chips.
本领域普通技术人员可以理解,上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。在硬件实施方式中,在以上描述中提及的功能模块/单元之间的划分不一定对应于物理组件的划分;例如,一个物理组件可以具有多个功能,或者一个功能或步骤可以由若干物理组件合作执行。某些组件或所有组件可以被实施为由处理器,如数字信号处理器或微处理器执行的软件,或者被实施为硬件,或者被实施为集成电路,如专用集成电路。这样的软件可以分布在计算机可读介质上,计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的,术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外,本领域普通技术人员公知的是,通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据,并且可包括任何信息递送介质。Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, functional modules/units in the systems, and devices can be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be composed of several physical components Components execute cooperatively. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). As is known to those of ordinary skill in the art, the term computer storage media includes both volatile and nonvolatile implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data flexible, removable and non-removable media. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, magnetic tape, magnetic disk storage or other magnetic storage devices, or may Any other medium used to store desired information and which can be accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and can include any information delivery media, as is well known to those of ordinary skill in the art .
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811535913.8A CN111324572A (en) | 2018-12-14 | 2018-12-14 | Chip adopting heterogeneous system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811535913.8A CN111324572A (en) | 2018-12-14 | 2018-12-14 | Chip adopting heterogeneous system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111324572A true CN111324572A (en) | 2020-06-23 |
Family
ID=71170183
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811535913.8A Pending CN111324572A (en) | 2018-12-14 | 2018-12-14 | Chip adopting heterogeneous system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111324572A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103150276A (en) * | 2011-11-28 | 2013-06-12 | 联发科技股份有限公司 | Dynamic configuration method and device |
CN107027162A (en) * | 2017-06-06 | 2017-08-08 | 福建强闽信息科技有限公司 | A kind of LoRa intelligent gateways and its application method, Internet of things system |
CN107645764A (en) * | 2017-08-25 | 2018-01-30 | 重庆慧联无限科技有限公司 | The multimode positioning label of low-power consumption |
CN207424866U (en) * | 2017-05-22 | 2018-05-29 | 无锡德思普科技有限公司 | A kind of data communication system between kernel based on heterogeneous multi-nucleus processor |
US20180206188A1 (en) * | 2017-01-18 | 2018-07-19 | Samsung Electronics Co., Ltd. | Method for internet of things communication and an electronic device thereof |
CN108494414A (en) * | 2018-05-10 | 2018-09-04 | 武汉中电物联科技有限公司 | Super low-power consumption multimode Internet of Things wireless module |
-
2018
- 2018-12-14 CN CN201811535913.8A patent/CN111324572A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103150276A (en) * | 2011-11-28 | 2013-06-12 | 联发科技股份有限公司 | Dynamic configuration method and device |
US20180206188A1 (en) * | 2017-01-18 | 2018-07-19 | Samsung Electronics Co., Ltd. | Method for internet of things communication and an electronic device thereof |
CN207424866U (en) * | 2017-05-22 | 2018-05-29 | 无锡德思普科技有限公司 | A kind of data communication system between kernel based on heterogeneous multi-nucleus processor |
CN107027162A (en) * | 2017-06-06 | 2017-08-08 | 福建强闽信息科技有限公司 | A kind of LoRa intelligent gateways and its application method, Internet of things system |
CN107645764A (en) * | 2017-08-25 | 2018-01-30 | 重庆慧联无限科技有限公司 | The multimode positioning label of low-power consumption |
CN108494414A (en) * | 2018-05-10 | 2018-09-04 | 武汉中电物联科技有限公司 | Super low-power consumption multimode Internet of Things wireless module |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110597559B (en) | Computing device and computing method | |
US20190026626A1 (en) | Neural network accelerator and operation method thereof | |
US20190187963A1 (en) | Memory access optimisation using per-layer computational mapping and memory allocation for cnn application | |
CN104778148A (en) | Dynamic reconfigurable embedded data co-processing platform based on FPGA (Field-Programmable Gate Array) and data processing method realized by using platform | |
TW201918883A (en) | High frequency wide memory system and logic die | |
CN113407479B (en) | A multi-core architecture with embedded FPGA and data processing method thereof | |
CN100574200C (en) | Smart Ethernet card with hardware acceleration | |
US20230367593A1 (en) | RISC-V-based Artificial Intelligence Inference Method and System | |
WO2023159957A1 (en) | Fpga-based full-stack network card task processing method and system | |
US20230085718A1 (en) | Neural network scheduling method and apparatus | |
CN109284192A (en) | Parameter configuration method and electronic device | |
CN112433596B (en) | Link width adjustment method, device, device and storage medium | |
CN107544937A (en) | A kind of coprocessor, method for writing data and processor | |
CN109491934A (en) | A kind of storage management system control method of integrated computing function | |
CN105426322A (en) | Data prefetching method and device | |
CN103093446B (en) | Multi-source image fusion device and method based on on-chip system of multiprocessor | |
EP4474999B1 (en) | Systems and methods for data management in storage devices | |
CN111324572A (en) | Chip adopting heterogeneous system | |
CN115189863B (en) | E-commerce transaction information management system based on block chain network architecture | |
CN103809987A (en) | Method for replacing FPGA IP program in SOC chip | |
CN106569968A (en) | Inter-array data transmission structure and scheduling method used for reconfigurable processor | |
US11792303B1 (en) | Fast clear memory of system memory | |
US20220334634A1 (en) | Systems and methods for reducing power consumption in embedded machine learning accelerators | |
CN104678815A (en) | Interface structure and configuration method of FPGA (field programmable gate array) chip | |
CN103247019A (en) | Reconfigurable Device Based on AdaBoost Algorithm for Object Detection |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200623 |