Multi Channel DMA Intel® FPGA IP for PCI Express User Guide

Download PDF
ID 683821
Date 4/11/2023
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A newer version of this document is available.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. Before You Begin 2. Introduction 3. Functional Description 4. Interface Overview 5. Parameters (H-Tile) 6. Parameters (P-Tile) (F-Tile) (R-Tile) 7. Designing with the IP Core 8. Software Programming Model 9. Registers 10. Troubleshooting/Debugging 11. Multi Channel DMA Intel FPGA IP for PCI Express User Guide Archives 12. Revision History for Multi Channel DMA Intel FPGA IP for PCI Express User Guide
1. Before You Begin x
1.1. Terms and Acronyms 1.2. Known Issues
2. Introduction x
2.1. Multi Channel DMA IP for PCI Express Features 2.2. Device Family Support 2.3. Recommended Speed Grades 2.4. Resource Utilization 2.5. Release Information
2.1. Multi Channel DMA IP for PCI Express Features x
2.1.1. Endpoint Mode 2.1.2. Root Port Mode
3. Functional Description x
3.1. Multi Channel DMA 3.2. Bursting Avalon-MM Master (BAM) 3.3. Bursting Avalon-MM Slave (BAS) 3.4. MSI Interrupt 3.5. Config Slave (CS) 3.6. Root Port Address Translation Table Enablement 3.7. Hard IP Reconfiguration Interface 3.8. Config TL Interface 3.9. Configuration Intercept Interface (EP Only) 3.10. Data Mover Only
3.1. Multi Channel DMA x
3.1.1. H2D Data Mover 3.1.2. D2H Data Mover 3.1.3. Descriptors 3.1.4. Avalon-MM PIO Master 3.1.5. Avalon-MM Write (H2D) and Read (D2H) Master 3.1.6. Avalon-ST Source (H2D) and Sink (D2H) 3.1.7. User MSI-X 3.1.8. User Functional Level Reset (FLR) 3.1.9. Control Registers
3.1.2. D2H Data Mover x
3.1.2.1. D2H Descriptor Fetch
3.1.3. Descriptors x
3.1.3.1. Support for Unaligned (or byte-aligned) Data Transfer 3.1.3.2. Metadata Support 3.1.3.3. MSI-X/Writeback
3.1.6. Avalon-ST Source (H2D) and Sink (D2H) x
3.1.6.1. Avalon-ST 1-Port Mode 3.1.6.2. Packet (File) Boundary 3.1.6.3. Metadata
3.4. MSI Interrupt x
3.4.1. Endpoint MSI Support through BAS 3.4.2. MSI Interrupt Controller
3.5. Config Slave (CS) x
3.5.1. 29 bit AVMM address format 3.5.2. 14 bit AVMM address format 3.5.3. Configuration Access Mechanism
3.10. Data Mover Only x
3.10.1. H2D Data Mover 3.10.2. D2H Data Mover 3.10.3. Application Specific Bits
3.10.1. H2D Data Mover x
3.10.1.1. H2D Descriptor Format (h2ddm_desc) 3.10.1.2. H2D Descriptor Completion Packet Format (h2ddm_desc_cmpl) 3.10.1.3. H2D Descriptor Status (h2ddm_desc_status)
3.10.2. D2H Data Mover x
3.10.2.1. D2H Descriptor Format (d2hdm_desc) 3.10.2.2. D2H Descriptor Status (d2hdm_desc_status)
4. Interface Overview x
4.1. Port List 4.2. Clocks 4.3. Resets 4.4. Multi Channel DMA 4.5. Bursting Avalon-MM Master (BAM) Interface 4.6. Bursting Avalon-MM Slave (BAS) Interface 4.7. MSI Interface 4.8. Config Slave Interface (RP only) 4.9. Hard IP Reconfiguration Interface 4.10. Config TL Interface 4.11. Configuration Intercept Interface (EP Only) 4.12. Data Mover Interface 4.13. Hard IP Status Interface
4.1. Port List x
4.1.1. Port List (H-Tile) 4.1.2. Port List (P-Tile) (F-Tile) (R-Tile)
4.2. Clocks x
4.2.1. F-Tile System PLL Reference Clock Requirements
4.4. Multi Channel DMA x
4.4.1. Avalon-MM PIO Master 4.4.2. Avalon-MM Write Master (H2D) 4.4.3. Avalon-MM Read Master (D2H) 4.4.4. Avalon-ST Source (H2D) 4.4.5. Avalon-ST Sink (D2H) 4.4.6. User Event MSI-X Interface 4.4.7. User Functional Level Reset (FLR) Interface
4.12. Data Mover Interface x
4.12.1. H2D Data Mover Interface 4.12.2. D2H Data Mover Interface
5. Parameters (H-Tile) x
5.1. IP Settings 5.2. Example Designs
5.1. IP Settings x
5.1.1. System Settings 5.1.2. MCDMA Settings 5.1.3. Device Identification Registers 5.1.4. Multifunction and SR-IOV System Settings Parameters [Endpoint Mode] 5.1.5. Configuration, Debug and Extension Options 5.1.6. PHY Characteristics 5.1.7. PCI Express / PCI Capabilities Parameters
5.1.7. PCI Express / PCI Capabilities Parameters x
5.1.7.1. Device 5.1.7.2. Link 5.1.7.3. MSI-X 5.1.7.4. Power Management 5.1.7.5. Vendor Specific Extended Capability (VSEC)
6. Parameters (P-Tile) (F-Tile) (R-Tile) x
6.1. Top-Level Settings 6.2. PCIe0 Settings 6.3. PCIe1 Settings 6.4. Example Designs
6.2. PCIe0 Settings x
6.2.1. Base Address Register 6.2.2. PCIe0 Configuration, Debug and Extension Options 6.2.3. PCIe0 Device Identification Registers 6.2.4. PCIe0 PCI Express / PCI Capabilities Parameters 6.2.5. MCDMA Settings
6.2.3. PCIe0 Device Identification Registers x
6.2.3.1. PCIe0 PF0 IDs 6.2.3.2. PCIe0 PF0 VF IDs
6.2.4. PCIe0 PCI Express / PCI Capabilities Parameters x
6.2.4.1. PCIe0 Device 6.2.4.2. PCIe0 Link 6.2.4.3. PCIe0 MSI-X 6.2.4.4. PCIe0 DEV SER 6.2.4.5. PCIe0 PRS 6.2.4.6. PCIe0 ACS Capabilities 6.2.4.7. PCIe0 ATS 6.2.4.8. PCIe0 TPH
6.2.4.6. PCIe0 ACS Capabilities x
6.2.4.6.1. PCIe0 ACS for Physical Functions 6.2.4.6.2. PCIe0 ACS for Virtual Functions
6.2.4.7. PCIe0 ATS x
6.2.4.7.1. PCIe0 ATS for Physical Functions 6.2.4.7.2. PCIe0 ATS for Virtual Functions
6.2.4.8. PCIe0 TPH x
6.2.4.8.1. PCIe0 TPH for Physical Functions 6.2.4.8.2. PCIe0 TPH for Virtual Functions
6.3. PCIe1 Settings x
6.3.1. PCIe1 Configuration, Debug and Extension Options
7. Designing with the IP Core x
7.1. Generating the IP Core 7.2. Simulating the IP Core 7.3. IP Core Generation Output - Intel® Quartus® Prime Pro Edition 7.4. Systems Integration and Implementation
7.4. Systems Integration and Implementation x
7.4.1. Required Supporting IP
8. Software Programming Model x
8.1. Multi Channel DMA IP Custom Driver 8.2. Multi Channel DMA IP DPDK Poll-Mode based Driver 8.3. Multi Channel DMA IP Kernel Mode Network Device Driver
8.1. Multi Channel DMA IP Custom Driver x
8.1.1. Architecture vfio-pci ifc_uio libmqdma Sample application 8.1.2. libmqdma library details 8.1.3. Application 8.1.4. Software Flow 8.1.5. API Flow 8.1.6. libmqdma Library API List 8.1.7. Request Structures
8.1.2. libmqdma library details x
8.1.2.1. Channel Initialization 8.1.2.2. Descriptor Memory Management 8.1.2.3. Descriptor Completion Status Update
8.1.5. API Flow x
8.1.5.1. Single Descriptor Load and Submit 8.1.5.2. Multiple Descriptor Load and Submit
8.1.6. libmqdma Library API List x
8.1.6.1. ifc_api_start 8.1.6.2. ifc_mcdma_port_by_name 8.1.6.3. ifc_qdma_device_get 8.1.6.4. ifc_num_channels_get 8.1.6.5. ifc_qdma_channel_get 8.1.6.6. ifc_qdma_acquire_channels 8.1.6.7. ifc_qdma_release_all_channels 8.1.6.8. ifc_qdma_device_put 8.1.6.9. ifc_qdma_channel_put 8.1.6.10. ifc_qdma_completion_poll 8.1.6.11. ifc_qdma_request_start 8.1.6.12. ifc_qdma_request_prepare 8.1.6.13. ifc_qdma_descq_queue_batch_load 8.1.6.14. ifc_qdma_request_submit 8.1.6.15. ifc_qdma_pio_read32 8.1.6.16. ifc_qdma_pio_write32 8.1.6.17. ifc_qdma_pio_read64 8.1.6.18. ifc_qdma_pio_write64 8.1.6.19. ifc_qdma_pio_read128 8.1.6.20. ifc_qdma_pio_write128 8.1.6.21. ifc_qdma_pio_read256 8.1.6.22. ifc_qdma_pio_write256 8.1.6.23. ifc_request_malloc 8.1.6.24. ifc_request_free 8.1.6.25. ifc_app_stop 8.1.6.26. ifc_qdma_poll_init 8.1.6.27. ifc_qdma_poll_add 8.1.6.28. ifc_qdma_poll_wait 8.1.6.29. ifc_mcdma_port_by_name
8.2. Multi Channel DMA IP DPDK Poll-Mode based Driver x
8.2.1. Architecture 8.2.2. MCDMA Poll Mode Driver 8.2.3. DPDK based application 8.2.4. Request Structures 8.2.5. Software Flow 8.2.6. API Flow 8.2.7. API List
8.2.2. MCDMA Poll Mode Driver x
8.2.2.1. Completion Status Update 8.2.2.2. Metadata Support 8.2.2.3. User MSI-X
8.3. Multi Channel DMA IP Kernel Mode Network Device Driver x
8.3.1. Architecture 8.3.2. Driver Information 8.3.3. Software Flow
8.3.2. Driver Information x
8.3.2.1. Device Management 8.3.2.2. Channel Management 8.3.2.3. Descriptor Memory Management 8.3.2.4. Completions Management 8.3.2.5. ethtool support 8.3.2.6. debugfs support 8.3.2.7. SRIOV Support
8.3.3. Software Flow x
8.3.3.1. Host to Device Flow 8.3.3.2. Device to Host Flow
9. Registers x
9.1. Queue Control (QCSR) 9.2. MSI-X Memory Space 9.3. Control Register (GCSR)
10. Troubleshooting/Debugging x
10.1. Debug Toolkit
10.1. Debug Toolkit x
10.1.1. Overview 10.1.2. Enabling the Debug Toolkit 10.1.3. Launching the Debug Toolkit 10.1.4. Using the Debug Toolkit
10.1.4. Using the Debug Toolkit x
10.1.4.1. Main View 10.1.4.2. Toolkit Parameters 10.1.4.3. Channel Parameters 10.1.4.4. Eye Viewer 10.1.4.5. Link Inspector
10.1.4.4. Eye Viewer x
10.1.4.4.1. Running Eye Viewer in the P-Tile Debug Toolkit 10.1.4.4.2. Running Eye Viewer in the F-Tile Debug Toolkit
1. Before You Begin
2. Introduction
3. Functional Description
4. Interface Overview
5. Parameters (H-Tile)
6. Parameters (P-Tile) (F-Tile) (R-Tile)
7. Designing with the IP Core
8. Software Programming Model
9. Registers
10. Troubleshooting/Debugging
11. Multi Channel DMA Intel FPGA IP for PCI Express User Guide Archives
12. Revision History for Multi Channel DMA Intel FPGA IP for PCI Express User Guide

8.1.1. Architecture

The figure below shows the software architecture block diagram of MCDMA custom driver.

Figure 45. Block Level Software Architecture

In the above block diagram, dotted lines represent memory mapped I/O interface. The other two lines represent read and write operations triggered by the device.

The Multi Channel DMA IP for PCI Express supports the following kernel based modules to expose the device to user space.
  • vfio-pci
  • UIO

These drivers do not perform any device management and indicate to the Operating System (OS) that the devices are being used by user space such that the OS does not perform any action (e.g. scanning the device etc.) on these devices.

vfio-pci

This is the secure kernel module, provided by kernel distribution. This module allows you to program the I/O Memory Managnment Unit (IOMMU). IOMMU is the hardware which helps to ensure memory safety in user space drivers. In case, if you are using Single Root I/O Virtualization (SR-IOV) , you can load vfio-pci and bind the device.

  • This module enables IOMMU programming and Function level reset (FLR)
  • To expose device Base Address Registers (BAR) to user space, vfio-pci enables ioctl
  • Supports MSI-X (Message Signal Interrupts extensions) interrupts
  • Kernel versions >= 5.7, supports the enablement of virtual functions by using sysfs interface.
If you are using kernel versions below 5.7, you have the following alternatives:
  • Use ifc_uio, which supports to enable VFs.
  • Apply the patch on kernel to enable virtual functions by using sysfs. It needs kernel rebuild.

ifc_uio

This is the alternative driver to vfio-pci, which doesn’t use IOMMU.

By using PCIe, sysfs, interrupt framework utilities this module reads allows the user space to access the device.

Like vfio-pci, this module can also be used from guest VM through the hypervisor. This driver allows the enablement/disablement of virtual functions. Once a virtual function is created, by default it binds to ifc_uio. Based on the requirement, you may unbind and bind to another driver.

Following are the functionalities supported by using this module:
  • Allows enablement/disablement of virtual functions by using sysfs interface.
  • Probes and exports channel BARs to libmqdma
  • Supports Interrupt notification/clearing

libmqdma

This is a user-space library used by the application to access the PCIe device.
  • This library has the APIs to access the MCDMA IP design and you can develop your application using this API.
  • It features calls for allocation, release and reset of the channels
  • libmqdma supports accessing the devices binded by UIO or Virtual Function I/O (VFIO ).
The libmqdma supports two user space drivers.
  • uio
  • vfio-pci

You can tune these options from the make file.

In case of UIO, ifc_uio driver reads the BAR register info by using sysfs and register MSI-X info by using eventfds.

In case of VFIO, user space uses IOCTL command to read BAR registers, MSIX information and programming of IOMMU table.

Typically, when an application is running in a virtualized environment, you bind the device to vfio-pci module and libmqdma can access the device by using ioctl . Currently, the support of UIO and VFIO can be switched from common.mk file. UIO is enabled by default.

Sample application

This application uses the APIs from libmqdma and takes the following command line arguments as the input.
  • Total message sizes/ time duration
  • Packet size per descriptor
  • Write/Read
  • Completion reporting method
  • Number of channels

It runs multiple threads for accessing the DMA channel. It also has performance measuring capabilities. Based on the number threads you are using and number of channels you are processing, queues are scheduled on threads.

Level Two Title