Arria V Avalon-ST Interface for PCIe Solutions User Guide

Download PDF
ID 683733
Date 6/03/2020
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents
Document Table of Contents x
1. Datasheet 2. Getting Started with the Arria V Hard IP for PCI Express 3. Parameter Settings 4. Interfaces and Signal Descriptions 5. Registers 6. Interrupts 7. Error Handling 8. IP Core Architecture 9. Transaction Layer Protocol (TLP) Details 10. Throughput Optimization 11. Design Implementation 12. Additional Features 13. Hard IP Reconfiguration 14. Transceiver PHY IP Reconfiguration 15. Testbench and Design Example 16. Debugging A. Transaction Layer Packet (TLP) Header Formats B. Lane Initialization and Reversal C. Document Revision History
1. Datasheet x
1.1. Arria V Avalon-ST Interface for PCIe Datasheet 1.2. Features 1.3. Release Information 1.4. Device Family Support 1.5. Configurations 1.6. Example Designs 1.7. Debug Features 1.8. IP Core Verification 1.9. Performance and Resource Utilization 1.10. Recommended Speed Grades 1.11. Creating a Design for PCI Express
1.8. IP Core Verification x
1.8.1. Compatibility Testing Environment
2. Getting Started with the Arria V Hard IP for PCI Express x
2.1. Qsys Design Flow 2.2. Modifying the Example Design 2.3. Using the IP Catalog To Generate Your Arria V Hard IP for PCI Express as a Separate Component
2.1. Qsys Design Flow x
2.1.1. Generating the Testbench 2.1.2. Simulating the Example Design 2.1.3. Generating Synthesis Files 2.1.4. Understanding the Files Generated 2.1.5. Understanding Physical Placement of the PCIe IP Core 2.1.6. Compiling the Design in the Quartus® Prime Software
3. Parameter Settings x
3.1. Avalon-ST System Settings 3.2. Link Capabilities 3.3. Port Function Parameters Shared Across All Port Functions 3.4. Port Function Parameters Defined Separately for All Port Functions
3.3. Port Function Parameters Shared Across All Port Functions x
3.3.1. Device Capabilities 3.3.2. Error Reporting 3.3.3. Link Capabilities 3.3.4. Slot Capabilities 3.3.5. Power Management
3.4. Port Function Parameters Defined Separately for All Port Functions x
3.4.1. Base Address Register (BAR) and Expansion ROM Settings 3.4.2. Base and Limit Registers for Root Ports 3.4.3. Device Identification Registers for Function <n> 3.4.4. Func <n> Device 3.4.5. Func <n> Link 3.4.6. Func <n> MSI and MSI-X Capabilities 3.4.7. Func <n> Legacy Interrupt
4. Interfaces and Signal Descriptions x
4.1. Arria V Hard IP for PCI Express with Avalon-ST Interface to the Application Layer 4.2. Clock Signals 4.3. Reset Signals 4.4. Hard IP Status 4.5. Error Signals 4.6. ECRC Forwarding 4.7. Interrupts for Endpoints 4.8. Interrupts for Root Ports 4.9. Completion Side Band Signals 4.10. Transaction Layer Configuration Space Signals 4.11. LMI Signals 4.12. Power Management Signals 4.13. Physical Layer Interface Signals
4.1. Arria V Hard IP for PCI Express with Avalon-ST Interface to the Application Layer x
4.1.1. Avalon‑ST RX Interface 4.1.2. Avalon-ST TX Interface
4.1.1. Avalon‑ST RX Interface x
4.1.1.1. Avalon-ST RX Component Specific Signals 4.1.1.2. Data Alignment and Timing for the 64‑Bit Avalon® -ST RX Interface 4.1.1.3. Data Alignment and Timing for the 128‑Bit Avalon‑ST RX Interface
4.1.2. Avalon-ST TX Interface x
4.1.2.1. Avalon-ST Packets to PCI Express TLPs 4.1.2.2. Data Alignment and Timing for the 64‑Bit Avalon-ST TX Interface 4.1.2.3. Data Alignment and Timing for the 128‑Bit Avalon‑ST TX Interface 4.1.2.4. Root Port Mode Configuration Requests
4.10. Transaction Layer Configuration Space Signals x
4.10.1. Configuration Space Register Access Timing 4.10.2. Configuration Space Register Access
4.13. Physical Layer Interface Signals x
4.13.1. Transceiver Reconfiguration 4.13.2. Serial Interface Signals 4.13.3. Hard IP Reconfiguration Interface 4.13.4. PIPE Interface Signals 4.13.5. Test Signals
4.13.2. Serial Interface Signals x
4.13.2.1. Physical Layout of Hard IP in Arria V Devices 4.13.2.2. Channel Placement in Arria V Devices
5. Registers x
5.1. Correspondence between Configuration Space Registers and the PCIe Specification 5.2. Type 0 Configuration Space Registers 5.3. Type 1 Configuration Space Registers 5.4. PCI Express Capability Structures 5.5. Intel-Defined VSEC Registers 5.6. CvP Registers 5.7. Uncorrectable Internal Error Mask Register 5.8. Uncorrectable Internal Error Status Register 5.9. Correctable Internal Error Mask Register 5.10. Correctable Internal Error Status Register
6. Interrupts x
6.1. Interrupts for Endpoints 6.2. Interrupts for Root Ports
6.1. Interrupts for Endpoints x
6.1.1. MSI Interrupts 6.1.2. MSI-X 6.1.3. Implementing MSI-X Interrupts
6.1.3. Implementing MSI-X Interrupts x
6.1.3.1. Legacy Interrupts
6.1.3.1. Legacy Interrupts x
6.1.3.1.1. Enabling MSI or Legacy Interrupts
7. Error Handling x
7.1. Physical Layer Errors 7.2. Data Link Layer Errors 7.3. Transaction Layer Errors 7.4. Error Reporting and Data Poisoning 7.5. Uncorrectable and Correctable Error Status Bits
8. IP Core Architecture x
8.1. Top-Level Interfaces 8.2. Transaction Layer 8.3. Data Link Layer 8.4. Physical Layer 8.5. Multi-Function Support
8.1. Top-Level Interfaces x
8.1.1. Avalon-ST Interface 8.1.2. Clocks and Reset 8.1.3. Local Management Interface (LMI Interface) 8.1.4. Hard IP Reconfiguration 8.1.5. Transceiver Reconfiguration 8.1.6. Interrupts 8.1.7. PIPE
8.2. Transaction Layer x
8.2.1. Configuration Space
9. Transaction Layer Protocol (TLP) Details x
9.1. Supported Message Types 9.2. Transaction Layer Routing Rules 9.3. Receive Buffer Reordering
9.1. Supported Message Types x
9.1.1. INTX Messages 9.1.2. Power Management Messages 9.1.3. Error Signaling Messages 9.1.4. Locked Transaction Message 9.1.5. Slot Power Limit Message 9.1.6. Vendor-Defined Messages 9.1.7. Hot Plug Messages
9.3. Receive Buffer Reordering x
9.3.1. Using Relaxed Ordering
10. Throughput Optimization x
10.1. Throughput of Posted Writes 10.2. Throughput of Non-Posted Reads
11. Design Implementation x
11.1. Making Analog QSF Assignments Using the Assignment Editor 11.2. Making Pin Assignments 11.3. Recommended Reset Sequence to Avoid Link Training Issues
12. Additional Features x
12.1. Configuration over Protocol (CvP) 12.2. Autonomous Mode 12.3. ECRC
12.2. Autonomous Mode x
12.2.1. Enabling Autonomous Mode 12.2.2. Enabling CvP Initialization
12.3. ECRC x
12.3.1. ECRC on the RX Path 12.3.2. ECRC on the TX Path
14. Transceiver PHY IP Reconfiguration x
14.1. Connecting the Transceiver Reconfiguration Controller IP Core 14.2. Transceiver Reconfiguration Controller Connectivity for Designs Using CvP
15. Testbench and Design Example x
15.1. Endpoint Testbench 15.2. Root Port Testbench 15.3. Test Driver Module 15.4. Root Port Design Example 15.5. Root Port BFM Overview 15.6. BFM Procedures and Functions 15.7. Setting Up Simulation
15.1. Endpoint Testbench x
15.1.1. Endpoint Testbench for SR-IOV
15.5. Root Port BFM Overview x
15.5.1. BFM Memory Map 15.5.2. Configuration Space Bus and Device Numbering 15.5.3. Configuration of Root Port and Endpoint 15.5.4. Issuing Read and Write Transactions to the Application Layer
15.6. BFM Procedures and Functions x
15.6.1. ebfm_barwr Procedure 15.6.2. ebfm_barwr_imm Procedure 15.6.3. ebfm_barrd_wait Procedure 15.6.4. ebfm_barrd_nowt Procedure 15.6.5. ebfm_cfgwr_imm_wait Procedure 15.6.6. ebfm_cfgwr_imm_nowt Procedure 15.6.7. ebfm_cfgrd_wait Procedure 15.6.8. ebfm_cfgrd_nowt Procedure 15.6.9. BFM Configuration Procedures 15.6.10. BFM Shared Memory Access Procedures 15.6.11. BFM Log and Message Procedures 15.6.12. Verilog HDL Formatting Functions
15.6.9. BFM Configuration Procedures x
15.6.9.1. ebfm_cfg_rp_ep Procedure 15.6.9.2. ebfm_cfg_decode_bar Procedure
15.6.10. BFM Shared Memory Access Procedures x
15.6.10.1. Shared Memory Constants 15.6.10.2. shmem_write Task 15.6.10.3. shmem_read Function 15.6.10.4. shmem_display Verilog HDL Function 15.6.10.5. shmem_fill Procedure 15.6.10.6. shmem_chk_ok Function
15.6.11. BFM Log and Message Procedures x
15.6.11.1. ebfm_display Verilog HDL Function 15.6.11.2. ebfm_log_stop_sim Verilog HDL Function 15.6.11.3. ebfm_log_set_suppressed_msg_mask Task 15.6.11.4. ebfm_log_set_stop_on_msg_mask Verilog HDL Task 15.6.11.5. ebfm_log_open Verilog HDL Function 15.6.11.6. ebfm_log_close Verilog HDL Function
15.6.12. Verilog HDL Formatting Functions x
15.6.12.1. himage1 15.6.12.2. himage2 15.6.12.3. himage4 15.6.12.4. himage8 15.6.12.5. himage16 15.6.12.6. dimage1 15.6.12.7. dimage2 15.6.12.8. dimage3 15.6.12.9. dimage4 15.6.12.10. dimage5 15.6.12.11. dimage6 15.6.12.12. dimage7
15.7. Setting Up Simulation x
15.7.1. Changing Between Serial and PIPE Simulation 15.7.2. Using the PIPE Interface for Gen1 and Gen2 Variants 15.7.3. Viewing the Important PIPE Interface Signals 15.7.4. Disabling the Scrambler for Gen1 and Gen2 Simulations 15.7.5. Disabling 8B/10B Encoding and Decoding for Gen1 and Gen2 Simulations 15.7.6. Changing between the Hard and Soft Reset Controller
16. Debugging x
16.1. Hardware Bring-Up Issues 16.2. Link Training 16.3. Use Third-Party PCIe Analyzer 16.4. BIOS Enumeration Issues
16.2. Link Training x
16.2.1. Link Hangs in L0 State
A. Transaction Layer Packet (TLP) Header Formats x
A.1. TLP Packet Formats without Data Payload A.2. TLP Packet Formats with Data Payload
C. Document Revision History x
C.1. Document Revision History of the Arria V Avalon® Streaming (Avalon-ST) Interface for PCIe* Solutions User Guide
1. Datasheet
2. Getting Started with the Arria V Hard IP for PCI Express
3. Parameter Settings
4. Interfaces and Signal Descriptions
5. Registers
6. Interrupts
7. Error Handling
8. IP Core Architecture
9. Transaction Layer Protocol (TLP) Details
10. Throughput Optimization
11. Design Implementation
12. Additional Features
13. Hard IP Reconfiguration
14. Transceiver PHY IP Reconfiguration
15. Testbench and Design Example
16. Debugging
A. Transaction Layer Packet (TLP) Header Formats
B. Lane Initialization and Reversal
C. Document Revision History

10. Throughput Optimization

The PCI Express Base Specification defines a flow control mechanism to ensure efficient transfer of TLPs.

Each transmitter, the write requester in this case, maintains a credit limit register and a credits consumed register. The credit limit register is the sum of all credits received by the receiver, the write completer in this case. The credit limit register is initialized during the flow control initialization phase of link initialization and then updated during operation by Flow Control (FC) Update DLLPs. The credits consumed register is the sum of all credits consumed by packets transmitted. Separate credit limit and credits consumed registers exist for each of the six types of Flow Control:

  • Posted Headers
  • Posted Data
  • Non-Posted Headers
  • Non-Posted Data
  • Completion Headers
  • Completion Data

Each receiver also maintains a credit allocated counter which is initialized to the total available space in the RX buffer (for the specific Flow Control class) and then incremented as packets are pulled out of the RX buffer by the Application Layer. The value of this register is sent as the FC Update DLLP value.

Figure 69. Flow Control Update Loop

The PCIe Hard IP maintains its own flow control logic, including a credit consumed register, and ensures that no TLP is sent that would use more credits than are available for that type of TLP. If you want optimum performance and granularity, you can maintain your own credit consumed register and flow control gating logic for each credit category (Header/Data, Posted/Non-posted/Completion). This allows you to halt the transmission of TLPs for a category that is out of credits, while still allowing TLP transmission for categories that have sufficient credits.

The following steps describe the Flow Control Update loop. The corresponding numbers in the figure show the general area to which they correspond.

  1. When the Application Layer has a packet to transmit, the number of credits required is calculated. If the required credits are less than or equal to the current value of available credits (credit limit - credits consumed so far), then the packet can be transmitted immediately. However, if the credit limit minus credits consumed is less than the required credits, then the packet must be held until the credit limit is increased to a sufficient value by an FC Update DLLP. This check is performed separately for the header and data credits; a single packet consumes only a single header credit.
  2. After the packet is selected for transmission the credits consumed register is incremented by the number of credits consumed by this packet. This increment happens for both the header and data credit consumed registers.
  3. The packet is received at the other end of the link and placed in the RX buffer.
  4. At some point the packet is read out of the RX buffer by the Application Layer. After the entire packet is read out of the RX buffer, the credit allocated register can be incremented by the number of credits the packet has used. There are separate credit allocated registers for the header and data credits.
  5. The value in the credit allocated register is used to create an FC Update DLLP.
  6. After an FC Update DLLP is created, it arbitrates for access to the PCI Express link. The FC Update DLLPs are typically scheduled with a low priority; consequently, a continuous stream of Application Layer TLPs or other DLLPs (such as ACKs) can delay the FC Update DLLP for a long time. To prevent starving the attached transmitter, FC Update DLLPs are raised to a high priority under the following three circumstances:
    1. When the last sent credit allocated counter minus the amount of received data is less than MAX_PAYLOAD and the current credit allocated counter is greater than the last sent credit counter. Essentially, this means the data sink knows the data source has less than a full MAX_PAYLOAD worth of credits, and therefore is starving.
    2. When an internal timer expires from the time the last FC Update DLLP was sent, which is configured to 30 µs to meet the PCI Express Base Specification for resending FC Update DLLPs.
    3. When the credit allocated counter minus the last sent credit allocated counter is greater than or equal to 25% of the total credits available in the RX buffer, then the FC Update DLLP request is raised to high priority.

      After arbitrating, the FC Update DLLP that won the arbitration to be the next item is transmitted. In the worst case, the FC Update DLLP may need to wait for a maximum sized TLP that is currently being transmitted to complete before it can be sent.

  7. The original write requester receives the FC Update DLLP. The credit limit value is updated. If packets are stalled waiting for credits, they can now be transmitted.
Note: You must keep track of the credits consumed by the Application Layer.

Section Content
Throughput of Posted Writes
Throughput of Non-Posted Reads

Level Two Title