P-Tile Avalon® Streaming Intel® FPGA IP for PCI Express* User Guide

Download PDF
ID 683059
Date 12/04/2023
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A newer version of this document is available.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. About the P-tile Avalon® Intel® FPGA IPs for PCI Express 2. IP Architecture and Functional Description 3. Advanced Features 4. Interfaces 5. Parameters 6. Testbench 7. Troubleshooting/Debugging 8. P-tile Avalon® Streaming Intel FPGA IP for PCI Express* User Guide Archives 9. Document Revision History for the P-Tile Avalon® Streaming Intel® FPGA IP for PCI Express* User Guide A. Configuration Space Registers B. Root Port Enumeration C. Implementation of Address Translation Services (ATS) in Endpoint Mode D. Packets Forwarded to the User Application in TLP Bypass Mode E. Using the Avery BFM for Intel P-Tile PCI Express Gen4 x16 Simulations F. Bifurcated Endpoint Support for Independent Warm Resets G. Margin Masks for the P-Tile Avalon Streaming Intel FPGA IP for PCI Express
1. About the P-tile Avalon® Intel® FPGA IPs for PCI Express x
1.1. Overview 1.2. Features 1.3. Release Information 1.4. Device Family Support 1.5. Performance and Resource Utilization 1.6. IP Core and Design Example Support Levels
1.1. Overview x
1.1.1. Standards and Specifications Compliance
2. IP Architecture and Functional Description x
2.1. Architecture 2.2. Functional Description
2.1. Architecture x
2.1.1. Clock Domains 2.1.2. Refclk 2.1.3. Reset
2.1.3. Reset x
2.1.3.1. Independent PERST
2.2. Functional Description x
2.2.1. PMA/PCS 2.2.2. Data Link Layer Overview 2.2.3. Transaction Layer Overview
3. Advanced Features x
3.1. PCIe* Port Bifurcation and PHY Channel Mapping 3.2. Virtualization Support 3.3. TLP Bypass Mode
3.2. Virtualization Support x
3.2.1. SR-IOV Support 3.2.2. VirtIO Support
3.2.1. SR-IOV Support x
3.2.1.1. SR-IOV Supported Features List 3.2.1.2. Implementation 3.2.1.3. VF to PF Mapping 3.2.1.4. Function Level Reset (FLR)
3.2.1.2. Implementation x
3.2.1.2.1. VF Error Flag Interface (for x16/x8 Cores Only)
3.2.2. VirtIO Support x
3.2.2.1. VirtIO Supported Features List 3.2.2.2. Overview 3.2.2.3. Parameters 3.2.2.4. VirtIO PCI Configuration Access Interface 3.2.2.5. Registers
3.2.2.5. Registers x
3.2.2.5.1. VirtIO Common Configuration Capability Register (Address: 0x012) 3.2.2.5.2. VirtIO Common Configuration BAR Indicator Register (Address: 0x013) 3.2.2.5.3. VirtIO Common Configuration BAR Offset Register (Address: 0x014) 3.2.2.5.4. VirtIO Common Configuration Structure Length Register (Address 0x015) 3.2.2.5.5. VirtIO Notifications Capability Register (Address: 0x016) 3.2.2.5.6. VirtIO Notifications BAR Indicator Register (Address: 0x017) 3.2.2.5.7. VirtIO Notifications BAR Offset Register (Address: 0x018) 3.2.2.5.8. VirtIO Notifications Structure Length Register (Address: 0x019) 3.2.2.5.9. VirtIO Notifications Notify Off Multiplier Register (Address: 0x01A) 3.2.2.5.10. VirtIO ISR Status Capability Register (Address: 0x02F) 3.2.2.5.11. VirtIO ISR Status BAR Indicator Register (Address: 0x030) 3.2.2.5.12. VirtIO ISR Status BAR Offset Register (Address: 0x031) 3.2.2.5.13. VirtIO ISR Status Structure Length Register (Address: 0x032) 3.2.2.5.14. VirtIO Device Specific Capability Register (Address: 0x033) 3.2.2.5.15. VirtIO Device Specific BAR Indicator Register (Address: 0x034) 3.2.2.5.16. VirtIO Device Specific BAR Offset Register (Address 0x035 ) 3.2.2.5.17. VirtIO Device Specific Structure Length Register (Address: 0x036) 3.2.2.5.18. VirtIO PCI Configuration Access Capability Register (Address: 0x037) 3.2.2.5.19. VirtIO PCI Configuration Access BAR Indicator Register (Address: 0x038) 3.2.2.5.20. VirtIO PCI Configuration Access BAR Offset Register (Address: 0x039) 3.2.2.5.21. VirtIO PCI Configuration Access Structure Length Register (Address: 0x03A) 3.2.2.5.22. VirtIO PCI Configuration Access Data Register (Address: 0x03B)
3.3. TLP Bypass Mode x
3.3.1. Overview 3.3.2. Register Settings for the TLP Bypass Mode 3.3.3. User Avalon® -MM Interface 3.3.4. Avalon® -ST Interface
3.3.2. Register Settings for the TLP Bypass Mode x
3.3.2.1. TLPBYPASS_ERR_EN (Address 0x104194) 3.3.2.2. TLPBYPASS_ERR_STATUS (Address 0x104190)
3.3.4. Avalon® -ST Interface x
3.3.4.1. Configuration TLP 3.3.4.2. Transmit Interface 3.3.4.3. Receive Interface 3.3.4.4. Malformed TLP 3.3.4.5. ECRC
4. Interfaces x
4.1. Overview 4.2. Clocks and Resets 4.3. Serial Data Interface 4.4. Avalon-ST Interface 4.5. Hard IP Status Interface 4.6. Interrupt Interface 4.7. Error Interface 4.8. Hot Plug Interface (RP Only) 4.9. Power Management Interface 4.10. Configuration Output Interface 4.11. Configuration Intercept Interface (EP Only) 4.12. Hard IP Reconfiguration Interface 4.13. PHY Reconfiguration Interface 4.14. Page Request Service (PRS) Interface (EP Only)
4.2. Clocks and Resets x
4.2.1. Interface Clock Signals 4.2.2. Resets
4.2.2. Resets x
4.2.2.1. Interface Reset Signals 4.2.2.2. Function-Level Reset (FLR) Interface (EP Only)
4.4. Avalon-ST Interface x
4.4.1. TLP Header and Data Alignment for the Avalon-ST RX and TX Interfaces 4.4.2. Avalon® -ST RX Interface 4.4.3. Avalon® -ST RX Interface rx_st_ready Behavior 4.4.4. RX Flow Control Interface 4.4.5. Avalon® -ST TX Interface 4.4.6. Avalon® -ST TX Interface tx_st_ready Behavior 4.4.7. TX Flow Control Interface 4.4.8. Tag Allocation
4.4.8. Tag Allocation x
4.4.8.1. Completion Buffer Size
4.6. Interrupt Interface x
4.6.1. Legacy Interrupts 4.6.2. MSI 4.6.3. MSI-X
4.6.3. MSI-X x
4.6.3.1. Implementing MSI-X Interrupts
4.7. Error Interface x
4.7.1. Completion Timeout Interface 4.7.2. VF Error Flags Interface
4.12. Hard IP Reconfiguration Interface x
4.12.1. Address Map for the User Avalon-MM Interface 4.12.2. Configuration Registers Access
4.12.1. Address Map for the User Avalon-MM Interface x
4.12.1.1. User Avalon-MM Control Register (Offset 0x104068)
4.12.2. Configuration Registers Access x
4.12.2.1. Using Direct User Avalon-MM Interface (Byte Access) 4.12.2.2. Using the Debug Register Interface Access (Dword Access)
5. Parameters x
5.1. Top-Level Settings 5.2. Core Parameters
5.2. Core Parameters x
5.2.1. Avalon® Parameters 5.2.2. Base Address Registers 5.2.3. PCI Express and PCI Capabilities Parameters 5.2.4. Device Identification Registers 5.2.5. Configuration, Debug and Extension Options
5.2.3. PCI Express and PCI Capabilities Parameters x
5.2.3.1. Device Capabilities 5.2.3.2. VirtIO Parameters 5.2.3.3. Link Capabilities 5.2.3.4. Legacy Interrupt Pin Register 5.2.3.5. MSI Capabilities 5.2.3.6. MSI-X Capabilities 5.2.3.7. Slot Capabilities 5.2.3.8. Latency Tolerance Reporting (LTR) 5.2.3.9. Process Address Space ID (PASID) 5.2.3.10. Device Serial Number Capability 5.2.3.11. Page Request Service (PRS) 5.2.3.12. Access Control Service (ACS) Capabilities 5.2.3.13. Power Management 5.2.3.14. Vendor Specific Extended Capability (VSEC) Registers 5.2.3.15. TLP Processing Hints (TPH) 5.2.3.16. Address Translation Services (ATS) Capabilities
5.2.3.1. Device Capabilities x
5.2.3.1.1. Multifunction and SR-IOV System Parameters
6. Testbench x
6.1. Endpoint Testbench 6.2. Test Driver Module 6.3. Root Port BFM
6.3. Root Port BFM x
6.3.1. BFM Memory Map 6.3.2. Configuration Space Bus and Device Numbering 6.3.3. Configuration of Root Port and Endpoint 6.3.4. Issuing Read and Write Transactions to the Application Layer 6.3.5. BFM Procedures and Functions
6.3.5. BFM Procedures and Functions x
6.3.5.1. ebfm_barwr Procedure 6.3.5.2. ebfm_barwr_imm Procedure 6.3.5.3. ebfm_barrd_wait Procedure 6.3.5.4. ebfm_barrd_nowt Procedure 6.3.5.5. ebfm_cfgwr_imm_wait Procedure 6.3.5.6. ebfm_cfgwr_imm_nowt Procedure 6.3.5.7. ebfm_cfgrd_wait Procedure 6.3.5.8. ebfm_cfgrd_nowt Procedure 6.3.5.9. BFM Configuration Procedures 6.3.5.10. BFM Shared Memory Access Procedures 6.3.5.11. BFM Log and Message Procedures 6.3.5.12. Verilog HDL Formatting Functions
6.3.5.9. BFM Configuration Procedures x
6.3.5.9.1. ebfm_cfg_rp_ep Procedure 6.3.5.9.2. ebfm_cfg_decode_bar Procedure
6.3.5.10. BFM Shared Memory Access Procedures x
6.3.5.10.1. Shared Memory Constants 6.3.5.10.2. shmem_write Task 6.3.5.10.3. shmem_read Function 6.3.5.10.4. shmem_display Verilog HDL Function 6.3.5.10.5. shmem_fill Procedure 6.3.5.10.6. shmem_chk_ok Function
6.3.5.11. BFM Log and Message Procedures x
6.3.5.11.1. ebfm_display Verilog HDL Function 6.3.5.11.2. ebfm_log_stop_sim Verilog HDL Function 6.3.5.11.3. ebfm_log_set_suppressed_msg_mask Task 6.3.5.11.4. ebfm_log_set_stop_on_msg_mask Verilog HDL Task 6.3.5.11.5. ebfm_log_open Verilog HDL Function 6.3.5.11.6. ebfm_log_close Verilog HDL Function
6.3.5.12. Verilog HDL Formatting Functions x
6.3.5.12.1. himage1 6.3.5.12.2. himage2 6.3.5.12.3. himage4 6.3.5.12.4. himage8 6.3.5.12.5. himage16 6.3.5.12.6. dimage1 6.3.5.12.7. dimage2 6.3.5.12.8. dimage3 6.3.5.12.9. dimage4 6.3.5.12.10. dimage5 6.3.5.12.11. dimage6 6.3.5.12.12. dimage7
7. Troubleshooting/Debugging x
7.1. Hardware 7.2. Debug Toolkit
7.1. Hardware x
7.1.1. Debugging Link Training Issues 7.1.2. Debugging Data Transfer and Performance Issues
7.1.1. Debugging Link Training Issues x
7.1.1.1. Generic Tools and Utilities 7.1.1.2. Signal Tap II Logic Analyzer 7.1.1.3. Additional Debug Tools
7.1.2. Debugging Data Transfer and Performance Issues x
7.1.2.1. Advanced Error Reporting (AER) 7.1.2.2. Second-Level Debug Tools
7.2. Debug Toolkit x
7.2.1. Overview 7.2.2. Enabling the P-Tile Debug Toolkit 7.2.3. Launching the P-Tile Debug Toolkit 7.2.4. Using the P-Tile Debug Toolkit 7.2.5. Using the P-Tile Link Inspector
A. Configuration Space Registers x
A.1. Configuration Space Registers A.2. Configuration Space Registers for Virtualization A.3. Intel-Defined VSEC Capability Registers
A.1. Configuration Space Registers x
A.1.1. Register Access Definitions A.1.2. PCIe Configuration Header Registers A.1.3. PCI Express Capability Structures A.1.4. Physical Layer 16.0 GT/s Extended Capability Structure A.1.5. MSI-X Registers
A.2. Configuration Space Registers for Virtualization x
A.2.1. SR-IOV Virtualization Extended Capabilities Registers Address Map A.2.2. PCIe Configuration Registers for Each Virtual Function
A.2.2. PCIe Configuration Registers for Each Virtual Function x
A.2.2.1. Alternative Routing ID (ARI) Capability Structure A.2.2.2. TLP Processing Hint (TPH) Capability Structure A.2.2.3. Address Translation Services (ATS) Capability Structure A.2.2.4. Access Control Services (ACS) Capability Structure
A.2.2.1. Alternative Routing ID (ARI) Capability Structure x
A.2.2.1.1. ARI Enhanced Capability Header Register (Offset 0x0) A.2.2.1.2. ARI Capability and Control Register (Offset 0x4)
A.2.2.2. TLP Processing Hint (TPH) Capability Structure x
A.2.2.2.1. TPH Requester Enhanced Capability Header (Offset 0x0) A.2.2.2.2. TPH Requester Capability Register (Offset 0x4) A.2.2.2.3. TPH Requester Control Register (Offset 0x8)
A.2.2.3. Address Translation Services (ATS) Capability Structure x
A.2.2.3.1. ATS Enhanced Capability Header (Offset 0x0) A.2.2.3.2. ATS Capability Register and ATS Control Register (Offset 0x4)
A.2.2.4. Access Control Services (ACS) Capability Structure x
A.2.2.4.1. ACS Extended Capability Header (Offset 0x0) A.2.2.4.2. ACS Capability Register (Offset 0x4) A.2.2.4.3. ACS Control Register (Offset 0x6) A.2.2.4.4. Egress Control Vector (Offset 0x8)
A.3. Intel-Defined VSEC Capability Registers x
A.3.1. Intel-Defined VSEC Capability Header (Offset 00h) A.3.2. Intel-Defined Vendor Specific Header (Offset 04h) A.3.3. Intel Marker (Offset 08h) A.3.4. JTAG Silicon ID (Offset 0x0C - 0x18) A.3.5. User Configurable Device and Board ID (Offset 0x1C - 0x1D) A.3.6. General Purpose Control and Status Register (Offset 0x30) A.3.7. Uncorrectable Internal Error Status Register (Offset 0x34) A.3.8. Uncorrectable Internal Error Mask Register (Offset 0x38) A.3.9. Correctable Internal Error Status Register (Offset 0x3C) A.3.10. Correctable Internal Error Mask Register (Offset 0x40)
C. Implementation of Address Translation Services (ATS) in Endpoint Mode x
C.1. Sending Translated/Untranslated Requests C.2. Sending a Page Request Message from the Endpoint (EP) to the Root Complex (RC) C.3. Invalidating Requests/Completions
D. Packets Forwarded to the User Application in TLP Bypass Mode x
D.1. EP TLP Bypass Mode (Upstream) D.2. RC TLP Bypass Mode (Downstream)
E. Using the Avery BFM for Intel P-Tile PCI Express Gen4 x16 Simulations x
E.1. Overview E.2. Generate the PCIe PIO Example Design E.3. Integrate Avery BFMs (Root Complex) E.4. Configure the Avery BFM and Update the Simulation Script E.5. Compile and Simulate E.6. View the Results
E.2. Generate the PCIe PIO Example Design x
E.2.1. Avalon® -ST PCIe PIO Example Design
F. Bifurcated Endpoint Support for Independent Warm Resets x
F.1. PCI Express Resets F.2. P-tile PCIe Design Constraints
F.2. P-tile PCIe Design Constraints x
F.2.1. P-tile Dual-Endpoint System Configurations
G. Margin Masks for the P-Tile Avalon Streaming Intel FPGA IP for PCI Express x
G.1. Margin Masks Overview G.2. Nx5 Testing Efficiency G.3. Debug Toolkit Margin Methodology
G.2. Nx5 Testing Efficiency x
G.2.1. First Time Testing G.2.2. Spot Check Testing
G.3. Debug Toolkit Margin Methodology x
G.3.1. Debug Toolkit (DTK) Testing Process G.3.2. DTK Margin Evaluation Based on Margin Mask Values
1. About the P-tile Avalon® Intel® FPGA IPs for PCI Express
2. IP Architecture and Functional Description
3. Advanced Features
4. Interfaces
5. Parameters
6. Testbench
7. Troubleshooting/Debugging
8. P-tile Avalon® Streaming Intel FPGA IP for PCI Express* User Guide Archives
9. Document Revision History for the P-Tile Avalon® Streaming Intel® FPGA IP for PCI Express* User Guide
A. Configuration Space Registers
B. Root Port Enumeration
C. Implementation of Address Translation Services (ATS) in Endpoint Mode
D. Packets Forwarded to the User Application in TLP Bypass Mode
E. Using the Avery BFM for Intel P-Tile PCI Express Gen4 x16 Simulations
F. Bifurcated Endpoint Support for Independent Warm Resets
G. Margin Masks for the P-Tile Avalon Streaming Intel FPGA IP for PCI Express

4.9. Power Management Interface

The P-Tile Avalon® Streaming Intel® FPGA IP for PCI Express* IP core supports the required D0 and D3 Power Management states. It does not support the optional D1 and D2 Power Management states. Software programs the Device into a D-state by writing to the Power Management Control and Status register in the PCI Power Management Capability Structure. The power management interface transmits the D-state to the Application Layer.

The correspondence between the device power states (D states) and link power states (L states) is as follows:

Table 71.  Relationship Between Device and Link Power States
Device Power State Link Power State
D0 L0
D1 (not supported) L1
D2 (not supported) L1
D3 L1, L2/L3 Ready

P-Tile does not support ASPM.

Endpoint D3 Entry
  1. The power management software must ensure that all outstanding non-posted requests have received their associated completions by polling that Transaction Pending bit in the Device Status register. Only then, it can put a function into the D3hot state by writing the appropriate value into the PowerState field of its Power Management Control and Status Register.
  2. The link is forced into the L1 state when all the enabled functions change to the D3hot state. In this state, the function can only initiate PME or PME_TO_ACK Messages and can only respond to configuration requests or the PME_Turn_Off Message.
  3. The power management software sends the PME_Turn_Off Message to the Endpoint to initiate power down. The delivery of the message TLP causes the link to transition to L0 and the Message will also be passed on to the Avalon® Streaming RX interface.
  4. The IP core automatically transmits a PME_TO_Ack Message to acknowledge the Turn Off request.
  5. When ready for the power removal D3cold state, the application logic in the Endpoint asserts p#_app_ready_entr_l23_i. The IP core will then send the PM_Enter_L23 DLLP and initiate the Link transition to L2/L3 Ready.
  6. The reference clock and power can finally be removed when the link has transitioned to the L23 Ready state.
Figure 38. Link Transition to L23 Ready

Endpoint D3 Exit

L1 exit:
  1. Initiated by Host : Power management software can write to the PowerState field of the function’s PMCSR register to change its PM state to D0. Alternatively, the host can initiate link retrain, link disable or hot reset for an L1 exit.
  2. Initiated by Device : For the Endpoint to exit the D3 state, the PME_en bit in the Power Management Control and Status (PMCSR) register needs to be set first. The Application Layer can then request a wake-up event by asserting apps_pm_xmt_pme_i, which causes the IP core to transmit a PM_PME Message. In addition, the IP core sets the PME_status bit in the PMCSR register to notify software that it has requested the wake-up. The PCIe Link states are indicated on the power management interface. The LTSSM state is indicated on the ltssm_state output.
Figure 39. Application Layer Requesting a Wake-up Event by Asserting apps_pm_xmt_pme_i

Endpoint L2 Exit

The L2 link state is not supported in P-tile. A power cycle for the FPGA is required to exit L2 state in this case.

Endpoint L3 Exit

A power cycle for the FPGA is required to exit L3 state.

Table 72.  Power Management Interface
Signal Name Direction Description Clock Domain EP/RP/BP
pm_state_o[2:0] O Indicates the current power state.
  • 3'b000 : L0 or IDLE
  • 3'b001 : L0s
  • 3'b010 : L1
  • 3'b011 : L2
  • 3'b100 : L3
  • Other values are invalid.
 coreclkout_hip EP/RP/BP

x16/x8: pm_dstate_o[31:0]

x4: pm_dstate_o[3:0]

O

Power management D-state for each function.

Each PF uses four consecutive bits. For example, pm_dstate_o[3:0] correspond to PF0, pm_dstate_o[7:4] correspond to PF1, and so on.

The x4 ports (Ports 2 and 3) only require four bits to service a single PF. The D-states are encoded as follows:
  • 4'b0001 : D0
  • 4'b0010 : D1
  • 4'b0100 : D2
  • 4'b1000 : D3
  • Other values are invalid.
Async EP/RP/BP

x16/x8: apps_pm_xmt_pme_i[7:0]

x4: NA

 I

The application logic asserts this signal for one cycle to wake up the Power Management Capability (PMC) state machine from a D1, D2, or D3 power state. Upon wake-up, the IP core sends a PM_PME message.

Each bit corresponds to a PF. For example, apps_pm_xmt_pme_i[0] is for PF0, apps_pm_xmt_pme_i[1] is for PF1, and so on.

coreclkout_hip EP/BP

x16/x8: app_ready_entr_l23_i

x4: NA

 I The application logic asserts this signal to indicate that it is ready to enter the L2/L3 Ready state. The app_ready_entr_l23_i signal is provided for applications that must control the L2/L3 Ready entry (in case certain tasks must be performed before going into L2/L3 Ready). The core delays sending PM_Enter_L23 (in response to PM_Turn_Off) until this signal becomes active. This is a level-sensitive signal. coreclkout_hip EP/BP
apps_pm_xmt_turnoff_i I This signal is a request from the Application Layer to generate a PM_Turn_Off message. The Application Layer must assert this signal for one clock cycle. The IP core does not return an acknowledgement or grant signal. The Application Layer must not pulse the same signal again until the previous message has been transmitted. coreclkout_hip RP
app_init_rst_i I The Application Layer uses this signal to request a hot reset to downstream devices. The hot reset request will be sent when a single-cycle pulse is applied to this pin. coreclkout_hip RP/BP
app_req_retry_en_i[x:0] where x = 7 for Port 0 and Port 1. For Port 2 and Port 3, this is a single-bit signal. I

When this signal is asserted, the PCIe Controller responds to Configuration TLPs with a CRS (Configuration Retry Status) if it has not already responded to a Configuration TLP with non-CRS status since the last reset. You can use this signal to hold off on enumeration. For Ports 2 and 3 which support Root Port mode only, this input is a single-bit signal.

Each bit corresponds to a PF. For example, app_req_retry_en_i[0] is for PF0, app_req_retry_en_i[1] is for PF1, and so on.

coreclkout_hip EP
p#_sys_aux_pwr_det_i I This is the Auxiliary Power Detected input (one signal for each port). It reports to the host software that auxiliary power (Vaux) is present. Refer to the Device Status Register in the PCI Express Capability Structure. Assertion of this signal will put the device into D3/L2 state instead of D3/L3 after L23_entry. Tie this signal to “0” as P-tile does not support the L2 state. coreclkout_hip EP/RP/BP
app_xfer_pending_i I This signal is only valid during L1 state. It is only used to make the device exit L1 and go back to L0. It is not able to prevent the device from entering L1. The PowerState field of the Power Management Control and Status Register takes higher precedence over this signal. It can trigger an L1 exit but the link will transition back to L1 again from L0 if the D3 state is not cleared. coreclkout_hip EP/RP/BP
Level Two Title