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A newer version of this document is available.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 Advantages Limitations 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

Advantages and Limitations of Arbitrary Precision Data Types

Advantages

The arbitrary precision data types have the following advantages over the use of standard C/C++ data types:

  • You can achieve narrower data paths and processing elements for various operations in the circuit.
  • The data types ensure that all operations are carried out in a size guaranteed not to lose any data. However, you can still lose data if you store data in a location where the data type is too narrow in size.

Limitations

AC Data Types

The AC data types have the following limitations:

  • Multipliers are limited to generating 512-bit results.
  • Dividers for ac_int data types are limited to a maximum of 64-bit unsigned or 63-bit signed.
  • You must initialize an ac_int variable before accessing it using the bit-select operator [] or bit-slice operations slc and set_slc. Using the bit-select operator or bit-slice operations on an uninitialized ac_int variable is an undefined behavior and can give you unexpected results. Assigning each bit explicitly using the [] operator or set_slc function does not count as initializing the ac_int variable.
  • Dividers for ac_fixed data types are limited to a maximum of 64-bits (unsigned or signed).
  • Creation of ac_fixed variables larger than 32 bits are supported only with the use of the bit_fill utility function.

    For example: 

    // Creating an ac_fixed with value set to 4294967298, which is larger than 2^32.
  
// Unsupported
ac_fixed<64, 64, false> v1 = ac_fixed<64, 64, false>(4294967298);

// Supported
// 4294967298 is 0b100000000000000000000000000000010 in binary
// Express that as two 32-bit numbers and use the bit_fill utility function.
const int vec_inp[2] = {0x00000001, 0x00000002};
ac_fixed<64, 64, false> bit_fill_res;
bit_fill_res.bit_fill<2>(vec_inp);
  • The AC data types are not supported on the Red Hat Enterprise Linux* (RHEL) 7 operating system for emulation due to a bug in the glibc version bundled with RHEL 7.
  • You cannot template the ac_complex data type with the ap_float data type.
  • When using the bit_fill_hex() function inside a kernel, pass the input string to the kernel through a char buffer and not as a string buffer. In addition, hardware and simulation compile flows do not support using a string literal or passing the string directly to the function. The following are the supported and unsupported code patterns:

    Supported Patterns

    // Supported Pattern 1: Passing string as a char sycl::buffer to the kernel 
ac_int<140, false> supported_example1(queue &q) {
  ac_int<140, false> a; 
  std::string hex_string{"0x177632EE7E265080BD54FF0CE7EF42C12"}; 
  constexpr int N = 36; // size of hex_string 

  buffer<ac_int<140, false>, 1> inp1(&a, 1); 
  // Note: the N + 1 ensures that the null byte  
  //terminating the char array buffer is copied
  buffer<char, 1> inp2(hex_string.c_str(), range<1>(N + 1)); 
  
  q.submit([&](handler &h) { 
    accessor x(inp1, h, read_write);
    accessor y(inp2, h, read_only); 
    h.single_task<class D>([=] { 
      x[0].bit_fill_hex(&y[0]);
    }); 
  }); 
  q.wait(); 
  return a; 
}
    // Supported Pattern 2: Create a char array with the string literal. 
ac_int<140, false> supported_example2(queue &q) {
  ac_int<140, false> a; 
  buffer<ac_int<140, false>, 1> inp1(&a, 1); 

  q.submit([&](handler &h) { 
    accessor x(inp1, h, read_write); 
    h.single_task<class D>([=] {
      char str[36] = "0x177632EE7E265080BD54FF0CE7EF42C12"; 
      x[0].bit_fill_hex(str);
    }); 
  }); 
  q.wait(); 
  return a; 
}

    Unsupported Patterns

    // Unsupported Pattern 1 – Using a string Literal, will result in compilation error 
ac_int<140, false> unsupported_example1(queue& q) {
  { 
    ac_int<140, false> a; 
    buffer<ac_int<140, false>, 1> a_buff(&a, 1); 
   
    q.submit([&](handler &h) { 
      accessor a_acc {a_buff, h, write_only, no_init}; 
      h.single_task<class A>([=]() { 
        a_acc[0].bit_fill_hex("1141e98e8c51b7ac7ad387d7f8ee4f1b9"); 
      }); 
    }); 
    q.wait_and_throw(); 
    return a; 
  } 
}
    // Unsupported Pattern 2 – Passing the string to the kernel in a string sycl::buffer 
ac_int<140, false> unsupported_example2(queue& q) {
  {
    std::string str{"1141e98e8c51b7ac7ad387d7f8ee4f1b9"}; 
    ac_int<140, false> a; 
    buffer<std::string, 1> str_buff(&str, 1); 
    buffer<ac_int<140, false>, 1> a_buff(&a, 1); 
    
    q.submit([&](handler &h) {
      accessor str_acc {str_buff, h, read_only}; 
      accessor a_acc {a_buff, h, write_only, no_init}; 
  
      h.single_task<class B>([=]() { 
        a_acc[0].bit_fill_hex(str_acc[0].c_str());
      }); 
    }); 
    q.wait_and_throw(); 
    return a; 
  } 
}
ap_float Data Type

The ap_float data type has the following limitations:

  • While the floating-point optimization of converting into constants is performed for float and double data types, it is not performed for the ap_float data type.
  • A limited set of math functions is supported. For details, see Math Functions Supported by ap_float Data Type.
  • Constant initialization works only with the round-towards-zero (RZERO) rounding mode.
  • For emulation, the ap_float math library is not supported on the Red Hat Enterprise Linux* (RHEL) 7 operating system.
  • When computing A^B using ap_float's ihc_pown function, if B is an unsigned type T of size N bits and is equal to the maximum unsigned value, redefine B to be of size N+1 bits. Otherwise, results will be incorrect. For example: 
    // Sample Code: 
ap_float<8, 7> a = 2; 
ac_int<4, false> b = 15; // max value that this ac_int can hold 
… = ihc_pown(a , b); // !!! Will produce incorrect result 
    // Workaround: 
ap_float<8, 7> a = 2;
ac_int<5, false> b = 15; // Workaround 
… = ihc_pown(a , b); // Will produce correct result
Parent topic: Variable-Precision Integer and Floating-Point Support
Level Two Title