Developer Guide

FPGA Optimization Guide for Intel® oneAPI Toolkits

Download PDF
ID 767853
Date 12/16/2022
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A newer version of this document is available.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
FPGA Optimization Guide for Intel® oneAPI Toolkits
FPGA Optimization Guide for Intel® oneAPI Toolkits x
Introduction To FPGA Design Concepts Analyze Your Design Optimize Your Design FPGA Optimization Flags, Attributes, Pragmas, and Extensions Quick Reference Additional Information Document Revision History for the FPGA Optimization Guide for Intel® oneAPI Toolkits Notices and Disclaimers
Introduction To FPGA Design Concepts x
FPGA Architecture Overview Concepts of FPGA Hardware Design Methods of Hardware Design How Source Code Becomes a Custom Hardware Datapath Scheduling Mapping Parallelism Models to FPGA Hardware Memory Types
FPGA Architecture Overview x
Adaptive Logic Module (ALM) Lookup Table (LUT) Register Digital Signal Processing (DSP) Block Random Access Memory (RAM) Blocks
Concepts of FPGA Hardware Design x
Maximum Frequency (fMAX) Latency Pipelining Throughput Datapath Control Path Occupancy
How Source Code Becomes a Custom Hardware Datapath x
Mapping Source Code Instructions to Hardware Mapping Arrays and Their Accesses to Hardware
Scheduling x
Dynamic Scheduling Clustering the Datapath Handshaking Between Clusters
Mapping Parallelism Models to FPGA Hardware x
Data Parallelism Task Parallelism
Data Parallelism x
Executing Independent Operations Simultaneously Pipelining
Memory Types x
Kernel Memory Global Memory
Analyze Your Design x
Analyze the FPGA Early Image Analyze the FPGA Image
Analyze the FPGA Early Image x
Review the FPGA Optimization Report Access HLD FPGA Reports in JSON Format
Review the FPGA Optimization Report x
Loop Analysis Bottlenecks Viewer Area Estimates System Viewer Kernel Memory Viewer Schedule Viewer
Analyze the FPGA Image x
Quartus (Static) Summary Intel® FPGA Dynamic Profiler for DPC++ System-level Profiling Using the Intercept Layer for OpenCL* Applications
Quartus (Static) Summary x
Timing Failures
Intel® FPGA Dynamic Profiler for DPC++ x
Measure Kernel Performance Instrument the Kernel Pipeline with Performance Counters (<span class='codeph'>-Xsprofile</span>) Obtain Profiling Data During Runtime Reduce Area Resource Use While Profiling Profiler Analyses of Example SYCL* Design Scenarios Limitations
Obtain Profiling Data During Runtime x
Invoke the Profiler Runtime Wrapper to Obtain Profiling Data Use Intel® VTune™ Profiler
Use Intel® VTune™ Profiler x
Interpret Performance Counter Data
System-level Profiling Using the Intercept Layer for OpenCL* Applications x
Set Up the Intercept Layer for OpenCL* Applications
Optimize Your Design x
Throughput Resource Use
Throughput x
Single Work-item Kernels NDRange Kernels Memory Accesses Pipes Host
Single Work-item Kernels x
Single Work-item Kernel Design Guidelines Loops Single-Cycle Floating-Point Accumulator for Single Work-Item Kernels
Loops x
Refactor the Loop-Carried Data Dependency Relax Loop-Carried Dependency Transfer Loop-Carried Dependency to Local Memory Minimize the Memory Dependencies for Loop Pipelining Unroll Loops Fuse Loops to Reduce Overhead and Improve Performance Optimize Loops With Loop Speculation Remove Loop Bottlenecks Shannonization to Improve FMAX/II Optimize Inner Loop Throughput Improve Loop Performance by Caching On-Chip Memory
Single-Cycle Floating-Point Accumulator for Single Work-Item Kernels x
Strategies for Inferring the Accumulator
Memory Accesses x
Load-Store Units Global Memory Accesses Optimization Perform Kernel Computations Using Local or Private Memory Local and Private Memory Accesses Optimization Annotating Unified Shared Memory Pointers Zero-Copy Memory Access Additional Recommendations
Load-Store Units x
Load-Store Unit Styles Load-Store Unit Modifiers Load-Store Unit Controls
Global Memory Accesses Optimization x
Global Memory Bandwidth Use Calculation Manual Partition of Global Memory Partitioning Buffers Across Different Memory Types (Heterogeneous Memory) Partitioning Buffers Across Memory Channels of the Same Memory Type Ignoring Dependencies Between Accessor Arguments Contiguous Memory Accesses Static Memory Coalescing
Host x
Multi-Threaded Host Application Utilizing Hardware Kernel Invocation Queue Double Buffering Host Utilizing Kernel Invocation Queue N-Way Buffering to Overlap Kernel Execution Prepinning Memory Simple Host-Device Streaming Buffered Host-Device Streaming
Double Buffering Host Utilizing Kernel Invocation Queue x
Applying Double-Buffering Using the Intercept Layer for OpenCL* Applications
Resource Use x
Data Types and Operations Kernel Variable Accesses
Data Types and Operations x
Optimize Floating-point Operation Avoid Expensive Functions Variable-Precision Integer and Floating-Point Support
Variable-Precision Integer and Floating-Point Support x
Advantages and Limitations of Arbitrary Precision Data Types Declare and Use the AC Data Types
Declare and Use the AC Data Types x
Declare the <span class='codeph'>ac_int</span> Data Type Declare the <span class='codeph'>ac_fixed</span> Data Type Declare the <span class='codeph'>ac_complex</span> Data Type Declare the <span class='codeph'>ap_float</span> Data Type
Declare the <span class='codeph'>ap_float</span> Data Type x
Conversion Rules for <span class='codeph'>ap_float</span> Operations with Explicit Precision Controls Comparison Operators Additional <span class='codeph'>ap_float</span> Functions Additional Data Types Provided by the <span class='codeph'>ap_float.hpp</span> Header File Quality of Results and the ap_float Data Type
FPGA Optimization Flags, Attributes, Pragmas, and Extensions x
Optimization Flags Kernel Attributes Kernel Controls Kernel Variables Memory Attributes Loop Directives Floating-Point Pragmas Latency Controls (Beta) System of Tasks Extension (<span class='codeph'>task_sequence</span>)
Optimization Flags x
Specify Schedule FMAX Target for Kernels (<span class='codeph'>-Xsclock=<clock target>) Disable Burst-Interleaving of Global Memory (<span class='codeph'>-Xsno-interleaving=<global_memory_type></span>) Force Ring Interconnect for Global Memory (<span class='codeph'>-Xsglobal-ring</span>) Force a Single Store Ring to Reduce Area (<span class='codeph'>-Xsforce-single-store-ring</span>) Force Fewer Read Data Reorder Units to Reduce Area (<span class='codeph'>-Xsnum-reorder</span>) Disable Hardware Kernel Invocation Queue (<span class='codeph'>-Xsno-hardware-kernel-invocation-queue</span>) Modify the Handshaking Protocol Between Clusters (<span class='codeph'>-Xshyper-optimized-handshaking</span>) Disable Automatic Fusion of Loops (<span class='codeph'>-Xsdisable-auto-loop-fusion</span>) Fuse Adjacent Loops With Unequal Trip Counts (<span class='codeph'>-Xsenable-unequal-tc-fusion</span>) Pipeline Loops in Non-task Kernels (<span class='codeph'>-Xsauto-pipeline</span>) Control Semantics of Floating-Point Operations (<span class='codeph'>-fp-model=<var><value></var> </span>) Modify the Rounding Mode of Floating-point Operations (<span class='codeph'>-Xsrounding=<rounding_type></span>) Global Control of Exit FIFO Latency of Stall-free Clusters (<span class='codeph'>-Xssfc-exit-fifo-type=<var><value></var> </span>) Enable the Read-Only Cache for Read-Only Accessors (<span class='codeph'>-Xsread-only-cache-size=<var><N></var>)</span> Control Hardware Implementation of the Supported Data Types and Math Operations (<span class='codeph'>-Xsdsp-mode=<var><option></var> </span>)
Kernel Attributes x
Specify Schedule FMAX Target for Kernels Specify a Workgroup Size Specify Number of SIMD WorkItems Omit Hardware that Generates and Dispatches Kernel IDs Omit Hardware to Support the <span class='codeph'>no_global_work_offset</span> Attribute in <span class='codeph'>parallel_for</span> Kernels Reduce Kernel Area and Latency
Kernel Controls x
Pipes Extension
Pipes Extension x
Key Properties of a Pipe Accessing Pipes The <span class='codeph'>pipe</span> Class and its Use I/O Pipes Characteristics of Pipes Restrictions of Pipes Guidelines for Designing Pipes Pipe and Atomic Fence
Loop Directives x
<span class='codeph'>disable_loop_pipelining</span> Attribute <span class='codeph'>initiation_interval</span> Attribute <span class='codeph'>ivdep</span> Attribute <span class='codeph'>loop_coalesce</span> Attribute <span class='codeph'>max_concurrency</span> Attribute <span class='codeph'>max_interleaving</span> Attribute <span class='codeph'>speculated_iterations</span> Attribute <span class='codeph'>unroll</span> Pragma Loop Fuse Functions and <span class='codeph'>nofusion</span> Attribute
System of Tasks Extension (<span class='codeph'>task_sequence</span>) x
Task Functions <span class='codeph'>task_sequence</span> Use Cases
Quick Reference x
Algorithmic C Data Types Floating Point Pragmas FPGA Accessor Properties FPGA Extensions FPGA Kernel Attributes FPGA Local Memory Function Latency Control Properties (Beta) FPGA LSU Controls FPGA Loop Directives FPGA Memory Attributes FPGA Optimization Flags Pipe API <span class='codeph'>task_sequence</span> Template Parameters and Function APIs
FPGA Optimization Guide for Intel® oneAPI Toolkits

Set Up the Intercept Layer for OpenCL* Applications

The Intercept Layer for OpenCL* Applications is available on GitHub* at https://github.com/intel/opencl-intercept-layer

To set up the Intercept Layer for OpenCL Applications, perform the following steps:

  1. Download Intercept Layer for OpenCL Applications version 2.2.1 or later from GitHub* at the following URL:

    https://github.com/intel/opencl-intercept-layer

  2. Build the Intercept Layer according to the instructions provided in How to Build the Intercept Layer for OpenCL* Applications.
  3. Ensure that you have set ENABLE_CLILOADER=1 when running cmake command. For example, run cmake -DENABLE_CLILOADER=1 ...
  4. Run the make command in the build directory. This step builds the cliloader loader utility.

    The cliloader executable should now exist in the <path to opencl-intercept-layer-master download>/<build dir>/cliloader/ directory.

  5. Add the directory to your PATH environment variable if you want to run multiple designs using cliloader.

    You can now pass your executables to cliloader to run them with the intercept layer. For details about the cliloader loader utility, see cliloader: A Intercept Layer for OpenCL* Applications Loader.

  6. Set cliloader and other Intercept Layer options.

    If you run multiple designs with the same options, set up a clintercept.conf file in your home directory. You can also set the options as environment variables by prefixing the option name with CLI_. For example, the DllName option can be set through the CLI_DllName environment variable. For a list of options, see Controls in How to Use the Intercept Layer for OpenCL Applications.

    Option/Variable Description
    DllName=$CMPLR_ROOT/linux/lib/libOpenCL.so The intercept layer must know where libOpenCL.so file from the original oneAPI build is.
    DevicePerformanceTiming=1 and DevicePerformanceTimelineLogging=1 These options print out runtime timeline information in the output of the executable run.
    ChromePerformanceTiming=1, ChromeCallLogging=1, ChromePerformanceTimingInStages=1 These variables set up the chrome tracer output and ensure the output has Queued, Submitted, and Execution stages.

These instructions set up the cliloader executable, which provides some flexibility by allowing for more control over when the layer is used or not used. If you prefer a local installation (for a single design) or a global installation (always ON for all designs), follow the instructions at How to Install the Intercept Layer for OpenCL Applications.

When you run the host executable with cliloader <executable> [executable args] command, the stderr output contains lines as shown in the following example: 

Device Timeline for clEnqueueWriteBuffer (enqueue 1) = 63267241140401 ns (queued), 
63267241149579 ns (submit), 63267241194205 ns (start), 63267242905519 ns (end)

These lines give the timeline information about a variety of oneAPI runtime calls. After the host executable finishes running, there is also a summary of the performance information for the run. After the executable runs, the data collected is placed in the CLIntercept_Dump directory, which is in the home directory by default. Its location can be adjusted using the DumpDir=<directory where you want the output files> cliloader option. The CLIntercept_Dump directory contains a file called clintercept_trace.json. You can load this JSON file in the Google* Chrome trace event profiling tool (chrome://tracing/) to visualize the timeline data collected by the run.

The following is a sample visualization of timeline data:

OpenCL Intercept Layer Full Example Trace

This visualization shows different calls executed through time. The X-axis is time, with the scale shown near the top of the page. The Y-axis shows different calls that are split up in several ways.

The left side (Y-axis) has two different types of numbers:

  • Numbers that contain a decimal point.
    • The part of the number before the decimal point orders the calls approximately by start time.
    • The part of the number after the decimal point represents the queue number the call was made in.
  • Numbers that do not contain a decimal point. These numbers represent the thread ID of the thread being run on in the operating system.

The colors in the trace represent different stages of execution:

  • Blue during the queued stage.
  • Yellow during the submitted stage.
  • Orange for the execution stage.

Identify gaps between consecutive execution stages and kernel runs to identify possible areas for optimization.

For an example use of Intercept Layer for OpenCL Applications, see Applying Double-Buffering Using the Intercept Layer for OpenCL* Applications.

Parent topic: System-level Profiling Using the Intercept Layer for OpenCL* Applications
Level Two Title