L- and H-tile Avalon® Streaming and Single Root I/O Virtualization (SR-IOV) Intel® FPGA IP for PCI Express* User Guide

Download PDF
ID 683111
Date 3/07/2022
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A newer version of this document is available.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. Introduction 2. Quick Start Guide 3. Interface Overview 4. Parameters 5. Designing with the IP Core 6. Block Descriptions 7. Interrupts 8. Registers 9. Testbench and Design Example 10. Troubleshooting and Observing the Link A. PCI Express Core Architecture B. TX Credit Adjustment Sample Code C. Root Port Enumeration D. Document Revision History
1. Introduction x
1.1. Avalon-ST Interface with Optional SR-IOV for PCIe Introduction 1.2. Features 1.3. Release Information 1.4. Device Family Support 1.5. Recommended Fabric Speed Grades 1.6. Performance and Resource Utilization 1.7. Transceiver Tiles 1.8. PCI Express IP Core Package Layout 1.9. Channel Availability
2. Quick Start Guide x
2.1. Design Components 2.2. Hardware and Software Requirements 2.3. Directory Structure 2.4. Generating the Design Example 2.5. Simulating the Design Example 2.6. Compiling the Design Example and Programming the Device 2.7. Installing the Linux Kernel Driver 2.8. Running the Design Example Application
3. Interface Overview x
3.1. Avalon-ST RX Interface 3.2. Avalon-ST TX Interface 3.3. TX Credit Interface 3.4. TX and RX Serial Data 3.5. Clocks 3.6. Function-Level Reset (FLR) Interface 3.7. Control Shadow Interface for SR-IOV 3.8. Configuration Extension Bus Interface 3.9. Hard IP Reconfiguration Interface 3.10. Interrupt Interfaces 3.11. Power Management Interface 3.12. Reset 3.13. Transaction Layer Configuration Interface 3.14. PLL Reconfiguration Interface 3.15. PIPE Interface (Simulation Only)
4. Parameters x
4.1. Stratix 10 Avalon-ST Settings 4.2. Multifunction and SR-IOV System Settings 4.3. Base Address Registers 4.4. Device Identification Registers 4.5. TPH/ATS Capabilities 4.6. PCI Express and PCI Capabilities Parameters 4.7. Configuration, Debug and Extension Options 4.8. PHY Characteristics 4.9. Example Designs
4.6. PCI Express and PCI Capabilities Parameters x
4.6.1. Device Capabilities 4.6.2. Link Capabilities 4.6.3. MSI and MSI-X Capabilities 4.6.4. Slot Capabilities 4.6.5. Power Management 4.6.6. Vendor Specific Extended Capability (VSEC)
5. Designing with the IP Core x
5.1. Generation 5.2. Simulation 5.3. IP Core Generation Output ( Intel® Quartus® Prime Pro Edition) 5.4. Integration and Implementation 5.5. Required Supporting IP Cores 5.6. Channel Layout and PLL Usage
5.2. Simulation x
5.2.1. Selecting Serial or PIPE Simulation
5.4. Integration and Implementation x
5.4.1. Clock Requirements 5.4.2. Reset Requirements
5.5. Required Supporting IP Cores x
5.5.1. Hard Reset Controller 5.5.2. TX PLL
6. Block Descriptions x
6.1. Interfaces 6.2. Errors reported by the Application Layer 6.3. Power Management 6.4. Transaction Ordering 6.5. RX Buffer
6.1. Interfaces x
6.1.1. TLP Header and Data Alignment for the Avalon-ST RX and TX Interfaces 6.1.2. Avalon-ST 256-Bit RX Interface 6.1.3. Avalon-ST 512-Bit RX Interface 6.1.4. Avalon-ST 256-Bit TX Interface 6.1.5. Avalon-ST 512-Bit TX Interface 6.1.6. TX Credit Interface 6.1.7. Interpreting the TX Credit Interface 6.1.8. Clocks 6.1.9. Update Flow Control Timer and Credit Release 6.1.10. Function-Level Reset (FLR) Interface 6.1.11. Resets 6.1.12. Interrupts 6.1.13. Control Shadow Interface for SR-IOV 6.1.14. Transaction Layer Configuration Space Interface 6.1.15. Configuration Extension Bus Interface 6.1.16. Hard IP Status Interface 6.1.17. Hard IP Reconfiguration 6.1.18. Power Management Interface 6.1.19. Serial Data Interface 6.1.20. PIPE Interface 6.1.21. Test Interface 6.1.22. PLL IP Reconfiguration 6.1.23. Message Handling
6.1.2. Avalon-ST 256-Bit RX Interface x
6.1.2.1. Avalon-ST RX Interface Three- and Four-Dword TLPs 6.1.2.2. Avalon-ST RX Interface rx_st_ready Deasserts for the 256-Bit Interface 6.1.2.3. Avalon-ST RX Interface rx_st_valid Deasserts 6.1.2.4. Avalon-ST RX Back-to-Back Transmission 6.1.2.5. Avalon-ST RX Interface Single-Cycle TLPs
6.1.3. Avalon-ST 512-Bit RX Interface x
6.1.3.1. PCIe TLP Layout Using the 512-Bit Interface
6.1.4. Avalon-ST 256-Bit TX Interface x
6.1.4.1. Avalon-ST TX Three- and Four-Dword TLPs 6.1.4.2. Avalon-ST TX Interface tx_st_ready Deassertion 6.1.4.3. Assertion and Deassertion of Avalon-ST TX Interface tx_st_valid
6.1.12. Interrupts x
6.1.12.1. MSI and Legacy Interrupts
6.1.23. Message Handling x
6.1.23.1. Endpoint Received Messages 6.1.23.2. Endpoint Transmitted Messages
6.2. Errors reported by the Application Layer x
6.2.1. Error Handling
6.3. Power Management x
6.3.1. Endpoint D3 Entry 6.3.2. End Point D3 Exit 6.3.3. Exit from D3 hot 6.3.4. Exit from D3 cold 6.3.5. Active State Power Management
6.4. Transaction Ordering x
6.4.1. TX TLP Ordering 6.4.2. RX TLP Ordering
6.5. RX Buffer x
6.5.1. Retry Buffer 6.5.2. Configuration Retry Status
7. Interrupts x
7.1. Interrupts for Endpoints
7.1. Interrupts for Endpoints x
7.1.1. MSI and Legacy Interrupts 7.1.2. MSI-X 7.1.3. Implementing MSI-X Interrupts 7.1.4. Legacy Interrupts
8. Registers x
8.1. Configuration Space Registers
8.1. Configuration Space Registers x
8.1.1. Register Access Definitions 8.1.2. PCI Configuration Header Registers 8.1.3. PCI Express Capability Structures 8.1.4. Intel Defined VSEC Capability Header 8.1.5. General Purpose Control and Status Register 8.1.6. Uncorrectable Internal Error Status Register 8.1.7. Uncorrectable Internal Error Mask Register 8.1.8. Correctable Internal Error Status Register 8.1.9. Correctable Internal Error Mask Register 8.1.10. SR-IOV Virtualization Extended Capabilities Registers Address Map
8.1.4. Intel Defined VSEC Capability Header x
8.1.4.1. Intel Defined Vendor Specific Header 8.1.4.2. Intel Marker 8.1.4.3. JTAG Silicon ID 8.1.4.4. User Configurable Device and Board ID
8.1.10. SR-IOV Virtualization Extended Capabilities Registers Address Map x
8.1.10.1. ARI Enhanced Capability Header 8.1.10.2. SR-IOV Enhanced Capability Registers 8.1.10.3. Initial VFs and Total VFs Registers 8.1.10.4. VF Device ID Register 8.1.10.5. Page Size Registers 8.1.10.6. VF Base Address Registers (BARs) 0-5 8.1.10.7. Secondary PCI Express Extended Capability Header 8.1.10.8. Lane Status Registers 8.1.10.9. Transaction Processing Hints (TPH) Requester Enhanced Capability Header 8.1.10.10. TPH Requester Capability Register 8.1.10.11. TPH Requester Control Register 8.1.10.12. Address Translation Services ATS Enhanced Capability Header 8.1.10.13. ATS Capability Register and ATS Control Register
9. Testbench and Design Example x
9.1. Endpoint Testbench 9.2. Test Driver Module 9.3. Root Port BFM Overview 9.4. BFM Procedures and Functions
9.1. Endpoint Testbench x
9.1.1. Endpoint Testbench for SR-IOV
9.3. Root Port BFM Overview x
9.3.1. BFM Memory Map 9.3.2. Configuration Space Bus and Device Numbering 9.3.3. Configuration of Root Port and Endpoint 9.3.4. Issuing Read and Write Transactions to the Application Layer
9.4. BFM Procedures and Functions x
9.4.1. ebfm_barwr Procedure 9.4.2. ebfm_barwr_imm Procedure 9.4.3. ebfm_barrd_wait Procedure 9.4.4. ebfm_barrd_nowt Procedure 9.4.5. ebfm_cfgwr_imm_wait Procedure 9.4.6. ebfm_cfgwr_imm_nowt Procedure 9.4.7. ebfm_cfgrd_wait Procedure 9.4.8. ebfm_cfgrd_nowt Procedure 9.4.9. BFM Configuration Procedures 9.4.10. BFM Shared Memory Access Procedures 9.4.11. BFM Log and Message Procedures 9.4.12. Verilog HDL Formatting Functions
9.4.9. BFM Configuration Procedures x
9.4.9.1. ebfm_cfg_rp_ep Procedure 9.4.9.2. ebfm_cfg_decode_bar Procedure
9.4.10. BFM Shared Memory Access Procedures x
9.4.10.1. Shared Memory Constants 9.4.10.2. shmem_write Task 9.4.10.3. shmem_read Function 9.4.10.4. shmem_display Verilog HDL Function 9.4.10.5. shmem_fill Procedure 9.4.10.6. shmem_chk_ok Function
9.4.11. BFM Log and Message Procedures x
9.4.11.1. ebfm_display Verilog HDL Function 9.4.11.2. ebfm_log_stop_sim Verilog HDL Function 9.4.11.3. ebfm_log_set_suppressed_msg_mask Task 9.4.11.4. ebfm_log_set_stop_on_msg_mask Verilog HDL Task 9.4.11.5. ebfm_log_open Verilog HDL Function 9.4.11.6. ebfm_log_close Verilog HDL Function
9.4.12. Verilog HDL Formatting Functions x
9.4.12.1. himage1 9.4.12.2. himage2 9.4.12.3. himage4 9.4.12.4. himage8 9.4.12.5. himage16 9.4.12.6. dimage1 9.4.12.7. dimage2 9.4.12.8. dimage3 9.4.12.9. dimage4 9.4.12.10. dimage5 9.4.12.11. dimage6 9.4.12.12. dimage7
10. Troubleshooting and Observing the Link x
10.1. Troubleshooting 10.2. PCIe Link Inspector Overview
10.1. Troubleshooting x
10.1.1. Simulation Fails To Progress Beyond Polling.Active State 10.1.2. Hardware Bring-Up Issues 10.1.3. Link Training 10.1.4. Link Hangs in L0 State 10.1.5. Use Third-Party PCIe Analyzer
10.2. PCIe Link Inspector Overview x
10.2.1. PCIe* Link Inspector Hardware
10.2.1. PCIe* Link Inspector Hardware x
10.2.1.1. Enabling the PCIe* Link Inspector 10.2.1.2. Launching the PCIe* Link Inspector 10.2.1.3. The PCIe* Link Inspector LTSSM Monitor 10.2.1.4. Accessing the Configuration Space and Transceiver Registers 10.2.1.5. Additional Status Commands
10.2.1.3. The PCIe* Link Inspector LTSSM Monitor x
10.2.1.3.1. LTSSM Monitor Registers 10.2.1.3.2. ltssm_save2file <file_name> 10.2.1.3.3. ltssm_file2console 10.2.1.3.4. ltssm_debug_check 10.2.1.3.5. ltssm_display <num_states> 10.2.1.3.6. ltssm_save_oldstates
10.2.1.4. Accessing the Configuration Space and Transceiver Registers x
10.2.1.4.1. PCIe* Link Inspector Commands 10.2.1.4.2. NPDME PLL and Channel Commands 10.2.1.4.3. Register Address Map
10.2.1.5. Additional Status Commands x
10.2.1.5.1. Displaying PLL Lock and Calibration Status Registers
A. PCI Express Core Architecture x
A.1. Transaction Layer A.2. Data Link Layer A.3. Physical Layer
D. Document Revision History x
D.1. Document Revision History for the Intel® L- and H-tile Avalon® Streaming and Single-Root I/O Virtualization (SR-IOV) IP for PCI Express* User Guide
1. Introduction
2. Quick Start Guide
3. Interface Overview
4. Parameters
5. Designing with the IP Core
6. Block Descriptions
7. Interrupts
8. Registers
9. Testbench and Design Example
10. Troubleshooting and Observing the Link
A. PCI Express Core Architecture
B. TX Credit Adjustment Sample Code
C. Root Port Enumeration
D. Document Revision History

7.1.1. MSI and Legacy Interrupts

The IP core generates single dword Memory Write TLPs to signal MSI interrupts on the PCI Express link. The Application Layer Interrupt Handler Module app_msi_req output port controls MSI interrupt generation. When asserted, it causes an MSI posted Memory Write TLP to be generated. The IP core constructs the TLP using information from the following sources:
  • The MSI Capability registers
  • The traffic class (app_msi_tc)
  • The message data specified by app_msi_num
To enable MSI interrupts, the Application Layer must first set the MSI enable bit. Then, it must disable legacy interrupts by setting the Interrupt Disable, bit 10 of the Command register.

Whenever the IP core gets an MSI request from the Interrupt Handler Module in the application layer (via the app_msi_req signal), it will generate an MSI TLP no matter the state of the MSI Enable or Bus Master Enable bits. The application logic must gate the MSI TLP based on the state of those bits.

Figure 53. Interrupt Handler Module in the Application Layer

The following figure illustrates a possible implementation of the Interrupt Handler Module with a per vector enable bit. Alternatively, the Application Layer could implement a global interrupt enable instead of this per vector MSI.

Figure 54. Example Implementation of the Interrupt Handler Block

There are 32 possible MSI messages. The number of messages requested by a particular component does not necessarily correspond to the number of messages allocated. For example, in the following figure, the Endpoint requests eight MSIs but is only allocated two. In this case, you must design the Application Layer to use only two allocated messages.

Figure 55. MSI Request Example

The following table describes three example implementations. The first example allocates all 32 MSI messages. The second and third examples only allocate 4 interrupts.

Table 52.  MSI Messages Requested, Allocated, and Mapped

MSI

Allocated

32

4

4

System Error

31

3

3

Hot Plug and Power Management Event

30

2

3

Application Layer

29:0

1:0

2:0

MSI interrupts generated for Hot Plug, Power Management Events, and System Errors always use Traffic Class 0. MSI interrupts generated by the Application Layer can use any Traffic Class. For example, a DMA that generates an MSI at the end of a transmission can use the same traffic control as was used to transfer data.

The following figure illustrates the interactions among MSI interrupt signals for the Root Port. The minimum latency possible between app_msi_req and app_msi_ack is one clock cycle. In this timing diagram app_msi_req can extend beyond app_msi_ack before deasserting. In other words, the earliest that app_msi_req can deassert is on the rising edge of clock cycle 5 (one cycle after app_msi_ack is asserted) as shown, but it can deassert in later clock cycles as well.

Figure 56. MSI Interrupt Signals Timing
By default, the Intel L-/H-tile Avalon Streaming IP for PCI Express provides support for MSI messages with 32-bit addresses. The 64-bit address support on MSI messages can be enabled with the pfX_pci_msi_64_bit_addr_cap parameter on the top-level entity. Use the following steps to enable the feature:
  1. Open the top-level file of the Intel L-/H-tile Avalon Streaming IP for PCI Express (<ip_name>/synth/<ip_name>.v).
  2. Locate the <ip_name>_altera_pcie_s10_hip_ast_<ip_version>_<auto_generated_postfix> and set the pfX_pci_msi_64_bit_addr_cap parameter to True for each physical function that requires the 64-bit address MSI support.
  3. Save the changes and recompile your project without regenerating the files.
Level Two Title