Visible to Intel only — GUID: wgx1490746636468
Ixiasoft
Product Discontinuance Notification
1. About the RapidIO Intel FPGA IP Core
2. Getting Started
3. Parameter Settings
4. Functional Description
5. Signals
6. Software Interface
7. Testbench
8. Platform Designer (Standard) Design Example
9. RapidIO Intel FPGA IP User Guide Archives
10. Document Revision History for the RapidIO Intel® FPGA IP User Guide
A. Initialization Sequence
B. Porting a RapidIO Design from the Previous Version of Software
2.1. Installing and Licensing Intel® FPGA IP Cores
2.2. Generating IP Cores
2.3. IP Core Generation Output ( Intel® Quartus® Prime Standard Edition)
2.4. RapidIO IP Core Testbench Files
2.5. Simulating IP Cores
2.6. Integrating Your IP Core in Your Design
2.7. Specifying Timing Constraints
2.8. Compiling the Full Design and Programming the FPGA
2.9. Instantiating Multiple RapidIO IP Cores
2.6.1. Calibration Clock
2.6.2. Dynamic Transceiver Reconfiguration Controller
2.6.3. Transceiver Settings
2.6.4. Adding Transceiver Analog Settings for Arria II GX, Arria II GZ, and Stratix IV GX Variations
2.6.5. External Transceiver PLL
2.6.6. Transceiver PHY Reset Controller for Intel® Arria® 10 and Intel® Cyclone® 10 GX Variations
2.9.1. Clock and Signal Requirements for Arria® V, Cyclone® V, and Stratix® V Variations
2.9.2. Clock and Signal Requirements for Arria II GX, Arria II GZ, Cyclone IV GX, and Stratix IV GX Variations
2.9.3. Correcting the Synopsys Design Constraints File to Distinguish RapidIO IP Core Instances
2.9.4. Sourcing Multiple Tcl Scripts for Variations other than Intel® Arria® 10 and Intel® Cyclone® 10 GX
6.2.1. Capability Registers (CARs)
6.2.2. Command and Status Registers (CSRs)
6.2.3. Maintenance Interrupt Control Registers
6.2.4. Receive Maintenance Registers
6.2.5. Transmit Maintenance Registers
6.2.6. Transmit Port-Write Registers
6.2.7. Receive Port-Write Registers
6.2.8. Input/Output Master Address Mapping Registers
6.2.9. Input/Output Slave Mapping Registers
6.2.10. Input/Output Slave Interrupts
6.2.11. Transport Layer Feature Register
6.2.12. Error Management Registers
6.2.13. Doorbell Message Registers
7.1. Reset, Initialization, and Configuration
7.2. Maintenance Write and Read Transactions
7.3. SWRITE Transactions
7.4. NWRITE_R Transactions
7.5. NWRITE Transactions
7.6. NREAD Transactions
7.7. Doorbell Transactions
7.8. Doorbell and Write Transactions With Transaction Order Preservation
7.9. Port-Write Transactions
7.10. Transactions Across the Avalon® -ST Pass-Through Interface
Visible to Intel only — GUID: wgx1490746636468
Ixiasoft
4.3.6.3. Receive Priority Threshold Values
The Physical layer implements the RapidIO specification deadlock prevention rules by accepting or retrying packets based on three programmable threshold levels, called Priority Threshold values. The algorithm uses the packet’s priority field value. The block determines whether to accept or retry a packet based on its priority, the threshold values, and the number of free blocks available in the receiver buffer, using the following rules:
- Packets of priority 0 (lowest priority) are retried if the number of available free 64-byte blocks is less than the Priority 0 Threshold.
- Packets of priority 1 are retried only if the number of available free 64-byte blocks is less than the Priority 1 Threshold.
- Packets of priority 2 are retried only if the number of available free 64-byte blocks is less than the Priority 2 Threshold.
- Packets of priority 3 (highest priority) are retried only if the receiver buffer is full.
The default threshold values are:
- Priority 2 Threshold = 10
- Priority 1 Threshold = 15
- Priority 0 Threshold = 20
You can specify other threshold values by turning off Auto-configured from receiver buffer size on the Physical Layer page in the RapidIO parameter editor.
The RapidIO parameter editor enforces the following constraints to ensure the threshold values increase monotonically by at least the maximum size of a packet (five buffers), as required by the deadlock prevention rules:
- Priority 2 Threshold > 9
- Priority 1 Threshold > Priority 2 Threshold + 4
- Priority 0 Threshold > Priority 1 Threshold + 4
- Priority 0 Threshold < Number of available buffers
The following figure shows sample threshold values in context to illustrate how they work together to enforce the deadlock prevention rules.
Figure 13. Receiver Threshold Levels