PCIE

---恢复内容开始---

高速差分总线、串行总线

每一条PCIe链路中只能连接两个设备这两个设备互为是数据发送端和数据接收端。PCIe链路可以由多条Lane组成，目前PCIe链路×1、×2、×4、×8、×16和×32宽度的PCIe链路，还有几乎不使用的×12链路。

PCIe总线规范	总线频率	单Lane的峰值带宽	编码方式	单个Lane带宽
1.x	1.25GHz	2.5GT/s	8/10b编码 每传输8比特有效数据，要附带两比特的校验位，实际要传输10比特数据。	250MB/s
2.x	2.5GHz	5GT/s	8/10b编码	500MB/s
3.0	4GHz	8GT/s	128/130b编码	1GB/s

在PCIe总线中，使用GT(Gigatransfer)计算PCIe链路的峰值带宽。GT是在PCIe链路上传递的峰值带宽，其计算公式为 总线频率×数据位宽×2。

在PCIe总线中，数据报文在接收和发送过程中，需要通过多个层次，包括事务层transaction（事务、学报、交易）、数据链路层和物理层。PCIe总线的层次结构如下。

1 PERST#信号全局复位信号

2 REFCLK+和REFCLK-信号 使用这组信号与处理器系统同步。

3 WAKE#信号  提交唤醒请求，使处理器系统重新为该PCIe设备提供主电源V_cc。

Virtex-7 FPGA Gen3
Integrated Block for
PCI Express v1.3

clock data recovery
(CDR) and differential signaling.

The key features of the Virtex-7 FPGA Gen3 Integrated Block for PCI Express (8.0 GT/s) core are:
• High-performance, highly flexible, scalable, and reliable, general-purpose I/O core
° Compliant with the PCI Express Base Specification, rev. 3.0
° Compatible with conventional PCI software model
• GTH transceivers
° 2.5 GT/s, 5.0 GT/s, and 8.0 GT/s line speeds
° 1-lane, 2-lane, 4-lane, and 8-lane operation
• Endpoint configuration
• Multiple Function and Single-Root I/O Virtualization in the Endpoint configuration
° 2 Physical Functions
° 6 Virtual Functions
• Standardized user interface(s)
° Compliant to AXI4-Stream
° Separate Requester, Completion, and Message interfaces
° Flexible Data Alignment
° Parity generation and checking on AXI4-Stream interfaces
° Easy-to-use packet-based protocol
° Full-duplex communication enabling
° Optional back-to-back transactions to enable greater link bandwidth utilization
° Support for flow control of data and discontinuation of an in-process transaction in
transmit direction
° Support for flow control of data in receive direction
• Compliant with PCI and PCI Express power management functions
• Optional Tag Management feature
• Maximum transaction payload of up to 1024 bytes
• End-to-End Cyclic Redundancy Check (ECRC)
• Advanced Error Reporting (AER)
• Multi-Vector MSI for up to 32 vectors and MSI-X
• Atomic Operations and TLP Processing Hints

使用Gen3 X8

TLP:transaction layer packet

TLP Receiver Engine:
 This state machine receives TLPs from the PCI-e core and determines
their validity and destination.
 Controls register writes.
 Passes Read Completion payloads to the DDR3 DMA (if implemented).
 Passes BAR0, BAR2, or BAR4 register Read Requests to the TLP
transmitter engine.
TLP Transmitter Engine:
 Controls register Reads and constructs the resulting Read Completion
packets.
 Transmit Arbiter
 Provides arbitration for access to the PCI-e Core transmit data link port
for the eight ADC channel DMAs.
 Provides arbitration for access to the PCI-e Core transmit data link port
for register Read Completions.
 Provides arbitration for access to the PCI-e Core transmit data link port
for DDR3 DMA Read Requests (if implemented).
Memory Aperture Interface:
 Provides BAR and address decoding and the generation of chip selects.
 Creates BAR0, BAR2, and BAR4 register Write Data buses for
distribution throughout the FPGA.
 Creates 16 BAR0 register Read buses, 16 BAR2 user register Read buses,
and a FLASH device BAR4 Read bus for distribution throughout the
FPGA.
 It provides the multiplexing of the various data Read sources for
submission to the TLP transmitter engine.
PCI-e Control and Status Registers:
 It provides status of the PCI-e link including link width and speed.
 It provides the Interrupt Flag and Enable register that latches the various
interrupts from all of the other modules. Interrupts are combined to
generate an interrupt control to the PCI-e Core.
 Converts AXI-Lite Master accesses into standard PCI-e read/write
tranactions to provide AXI access to the Global Control Registers