1. WHAT IS AXI?
AXI是一种高级可扩展接口，是ARM AMBA的一部分。
2. WHAT IS AMBA?
AMBA是高级微控制器总线架构，开放的片内互联总线标准。
3．AXI分类
AXI4/AXI4_ITE/AXI4_STREAM
Memory_mapped 存储器映射 就是读写数据时需要目标地址。
AXI4(AXI4_FULL):用于高性能存储映射。例如：AXI4读写DDR3。
AXI4_LITE:简单的，低吞吐率。比如说一些控制寄存器和状态寄存器。
AXI4_STREAM（ST）：高速的流数据通信。
AXI4(AXI4_FULL)和AXI4_LITE需要主机指定读写的地址。AXI4_STREAM（ST）属于流数据，不需要存储器映射不需要地址。
4.存储器映射（Memory_mapped）
主机在对从机进行读写操作时，要先指定一个目标地址（target address），在这个地址对应系统存储空间的地址，表示对该空间进行读写操作。
Memory-Mapped Protocols: In memory-mapped protocols (AXI3, AXI4, and
AXI4-Lite), all transactions involve the concept of transferring a targetaddress within a
system memory space and data.
AXI_Stream不属于存储器映射。
5.AXI4的优势
AXI的优势
生产力：
灵活性：AXI4（支持突发256）和AXI4_lite（不支持突发，1个数据）都属于存储器映射。AXI_ST不属于存储器映射，突发长度不受限制。
可获得性：
6.what is 互联？
7. five different channels
Both AXI4 and AXI4-Lite interfaces consist of five different channels:
• Read Address Channel
• Write Address Channel
• Read Data Channel
• Write Data Channel
• Write Response Channel
主机发送读地址，从机返回读数据
Data can move in both directions betweenthe master and slave simultaneously, and data
transfer sizes can vary. The limit in AXI4 is a bursttransaction of up to 256 data transfers.
AXI4-Lite allows only one data transfer pertransaction.
读操作：主机通过读地址通道向从机发出地址和控制，从机通过读数据通道发送读数据。根据突发长度大小返回不同个数的数据。
写操作：主机先通过写地址通道发出地址和控制信号。然后通过写数据通道发出数据，等从机接收到数据会发送response信号。
AXI4:
读写地址分离，支持同时读写。
需要一个单一地址然后进行256个数据突发。
As shown in the preceding figures, AXI4:
• Provides separate data and address connections forreads and writes, which allows
simultaneous, bidirectional data transfer.
• Requires a single address and then bursts up to 256words of data.
AXI4_LITE:
单次一个数据，不支持突发传输。
AXI4-Lite is similar to AXI4 with someexceptions: The most notable exception is that
bursting is not supported.
AXI4-Stream：
单通道传输数据类似AXI4的写数据通道，且写数据大小没有突发限制。
The AXI4-Stream protocol defines a single channel fortransmission of streaming data. The AXI4-Stream channel models the write data channel of AXI4.Unlike AXI4, AXI4-Stream
interfaces can burst an unlimited amount of data.
只关心数据流，不需要地址。
AXI4-StreamProtocol: Use the AXI4-Stream protocol for applications that typically
focus on a data-centric and data-flow paradigm where the concept of an addressis not
present or not required. Each AXI4-Stream acts as a single unidirectionalchannel with a
handshaking data flow.
InfrastructureIP: An infrastructure IP is a building block used to help assemble
systems. Infrastructure IP tends to be a generic IP that moves or transformsdata
around the system using general-purpose AXI4 interfaces and does not interpretdata.
Examples of infrastructure IP are:
° AXI Register slices (for pipelining)
° AXI FIFOs (for buffering/clock conversion)
° AXI Interconnect IP and AXI SmartConnect IP (forconnecting memory-mapped IP
together)
° AXI Direct Memory Access (DMA) engines (for memory-mappedto stream
conversion)
° AXI Performance Monitors and Protocol Checkers(for analysis and debug)
° AXI Verification IP (for simulation-basedverification and performance analysis)
These IP are useful for connecting IP together into a system, but are notgenerally
endpoints for data.
DMA常用于存储器映射和stream数据流的转换。
比如说ddr中存储了你的有效数据需要做fftfft IP的接口时stream流 ddr的数据是存储器映射，做操作需要DMA做数据转换。
A commonapproach is to build systems that combine AXI4-Stream and AXI
memory-mapped IP together. Often a DMA engine can be used to move streams inand out
of memory.
注意AXI Data widthconverter 和 AXI Direct MemoryAccess支持的AXI接口协议。
8.AXI SmartConnect IP and AXIInterconnect IP
两者都用于一些主从设备的互联（存储器映射类型）。在一些情况下, AXISmartConnect IP可以提供高带宽的链接以及低延迟。
The XilinxLogiCORE IP AXI Interconnect and LogiCORE IP AXI SmartConnect cores both connectone or more AXI memory-mapped master devices to one or more memory-mapped slavedevices; however, the SmartConnect is more tightly integrated into the Vivadodesign environment to automatically configure and adapt to connected AXI masterand slave IP with minimal user intervention. The AXI Interconnect can be used inall memory-mapped designs. There are certain cases for high bandwidthapplication where using a SmartConnect provides better optimization. The AXISmartConnect IP delivers the maximum system throughput at low latency bysynthesizing a low area custom interconnect that is optimized for importantinterfaces.The AXI Interconnect core IP (axi_interconnect)connects one or more AXI memory-mapped master devices to one or morememory-mapped slave devices。
今天先到这里。
参考资料：
IHI0022D
UG1037
正点原子ZYNQ嵌入式教程