OSchip
/
OpenBLAS


<!doctype html>
<html lang="en" class="no-js">
  <head>
    
      <meta charset="utf-8">
      <meta name="viewport" content="width=device-width,initial-scale=1">
      
      
        <link rel="canonical" href="https://openblas.net/docs/developers/">
      
      
        <link rel="prev" href="../extensions/">
      
      
        <link rel="next" href="../build_system/">
      
      
      <link rel="icon" href="../logo.svg">
      <meta name="generator" content="mkdocs-1.6.0, mkdocs-material-9.5.22">
    
    
        <title>Developer manual - OpenBLAS</title>
      
    
      <link rel="stylesheet" href="../assets/stylesheets/main.732c4fb1.min.css">
      
        
        <link rel="stylesheet" href="../assets/stylesheets/palette.06af60db.min.css">
      
      
        <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
        <link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Roboto:300,300i,400,400i,700,700i%7CRoboto+Mono:400,400i,700,700i&display=fallback">
        <style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
      
    
    <script>__md_scope=new URL("..",location),__md_hash=e=>[...e].reduce((e,_)=>(e<<5)-e+_.charCodeAt(0),0),__md_get=(e,_=localStorage,t=__md_scope)=>JSON.parse(_.getItem(t.pathname+"."+e)),__md_set=(e,_,t=localStorage,a=__md_scope)=>{try{t.setItem(a.pathname+"."+e,JSON.stringify(_))}catch(e){}}</script>
    
      
  </head>
  
  
    <body dir="ltr" data-md-color-scheme="default" data-md-color-primary="grey" data-md-color-accent="indigo">
  
    
    <input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
    <input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
    <label class="md-overlay" for="__drawer"></label>
    <div data-md-component="skip">
      
        
        <a href="#developer-manual" class="md-skip">
          Skip to content
        </a>
      
    </div>
    <div data-md-component="announce">
      
    </div>
    
    
<header class="md-header md-header--shadow" data-md-component="header">
  <nav class="md-header__inner md-grid" aria-label="Header">
    <a href=".." title="OpenBLAS" class="md-header__button md-logo" aria-label="OpenBLAS" data-md-component="logo">
      
  <img src="../logo.svg" alt="logo">

    </a>
    <label class="md-header__button md-icon" for="__drawer">
      
      <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3V6m0 5h18v2H3v-2m0 5h18v2H3v-2Z"/></svg>
    </label>
    <div class="md-header__title" data-md-component="header-title">
      <div class="md-header__ellipsis">
        <div class="md-header__topic">
          <span class="md-ellipsis">
            OpenBLAS
          </span>
        </div>
        <div class="md-header__topic" data-md-component="header-topic">
          <span class="md-ellipsis">
            
              Developer manual
            
          </span>
        </div>
      </div>
    </div>
    
      
      <label class="md-header__button md-icon" for="__search">
        
        <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.516 6.516 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5Z"/></svg>
      </label>
      <div class="md-search" data-md-component="search" role="dialog">
  <label class="md-search__overlay" for="__search"></label>
  <div class="md-search__inner" role="search">
    <form class="md-search__form" name="search">
      <input type="text" class="md-search__input" name="query" aria-label="Search" placeholder="Search" autocapitalize="off" autocorrect="off" autocomplete="off" spellcheck="false" data-md-component="search-query" required>
      <label class="md-search__icon md-icon" for="__search">
        
        <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.516 6.516 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5Z"/></svg>
        
        <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M20 11v2H8l5.5 5.5-1.42 1.42L4.16 12l7.92-7.92L13.5 5.5 8 11h12Z"/></svg>
      </label>
      <nav class="md-search__options" aria-label="Search">
        
        <button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
          
          <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M19 6.41 17.59 5 12 10.59 6.41 5 5 6.41 10.59 12 5 17.59 6.41 19 12 13.41 17.59 19 19 17.59 13.41 12 19 6.41Z"/></svg>
        </button>
      </nav>
      
    </form>
    <div class="md-search__output">
      <div class="md-search__scrollwrap" data-md-scrollfix>
        <div class="md-search-result" data-md-component="search-result">
          <div class="md-search-result__meta">
            Initializing search
          </div>
          <ol class="md-search-result__list" role="presentation"></ol>
        </div>
      </div>
    </div>
  </div>
</div>
    
    
  </nav>
  
</header>
    
    <div class="md-container" data-md-component="container">
      
      
      <main class="md-main" data-md-component="main">
        <div class="md-main__inner md-grid">
          
            
              <div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
                <div class="md-sidebar__scrollwrap">
                  <div class="md-sidebar__inner">
                    

<nav class="md-nav md-nav--primary" aria-label="Navigation" data-md-level="0">
  <label class="md-nav__title" for="__drawer">
    <a href=".." title="OpenBLAS" class="md-nav__button md-logo" aria-label="OpenBLAS" data-md-component="logo">
      
  <img src="../logo.svg" alt="logo">

    </a>
    OpenBLAS
  </label>
  
  <ul class="md-nav__list" data-md-scrollfix>
    
      
    <li class="md-nav__item">
      <a href=".." class="md-nav__link">
        
  
  <span class="md-ellipsis">
    Home
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../install/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    Install OpenBLAS
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../user_manual/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    User manual
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../extensions/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    Extensions
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item md-nav__item--active">
      
      <input class="md-nav__toggle md-toggle" type="checkbox" id="__toc">
      
      
        <label class="md-nav__link md-nav__link--active" for="__toc">
          
  
  <span class="md-ellipsis">
    Developer manual
  </span>
  

          <span class="md-nav__icon md-icon"></span>
        </label>
      
      <a href="./" class="md-nav__link md-nav__link--active">
        
  
  <span class="md-ellipsis">
    Developer manual
  </span>
  

      </a>
      
        
<nav class="md-nav md-nav--secondary" aria-label="Table of contents">
  
  
    <label class="md-nav__title" for="__toc">
      <span class="md-nav__icon md-icon"></span>
      Table of contents
    </label>
    <ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
      
        <li class="md-nav__item">
  <a href="#source-codes-layout" class="md-nav__link">
    <span class="md-ellipsis">
      Source codes Layout
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#optimizing-gemm-for-a-given-hardware" class="md-nav__link">
    <span class="md-ellipsis">
      Optimizing GEMM for a given hardware
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#run-openblas-test" class="md-nav__link">
    <span class="md-ellipsis">
      Run OpenBLAS Test
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#benchmarking" class="md-nav__link">
    <span class="md-ellipsis">
      Benchmarking
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#adding-autodetection-support-for-a-new-revision-or-variant-of-a-supported-cpu" class="md-nav__link">
    <span class="md-ellipsis">
      Adding autodetection support for a new revision or variant of a supported cpu
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#adding-dedicated-support-for-a-new-cpu-model" class="md-nav__link">
    <span class="md-ellipsis">
      Adding dedicated support for a new cpu model
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#adding-support-for-an-entirely-new-architecture" class="md-nav__link">
    <span class="md-ellipsis">
      Adding support for an entirely new architecture
    </span>
  </a>
  
</li>
      
    </ul>
  
</nav>
      
    </li>
  

    <li class="md-nav__item">
      <a href="../build_system/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    Build system
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../distributing/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    Redistributing OpenBLAS
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../ci/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    CI jobs
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../about/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    About
  </span>
  

      </a>
    </li>
  

    <li class="md-nav__item">
      <a href="../faq/" class="md-nav__link">
        
  
  <span class="md-ellipsis">
    FAQ
  </span>
  

      </a>
    </li>
  

  </ul>
</nav>
                  </div>
                </div>
              </div>
            
            
              <div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
                <div class="md-sidebar__scrollwrap">
                  <div class="md-sidebar__inner">
                    

<nav class="md-nav md-nav--secondary" aria-label="Table of contents">
  
  
    <label class="md-nav__title" for="__toc">
      <span class="md-nav__icon md-icon"></span>
      Table of contents
    </label>
    <ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
      
        <li class="md-nav__item">
  <a href="#source-codes-layout" class="md-nav__link">
    <span class="md-ellipsis">
      Source codes Layout
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#optimizing-gemm-for-a-given-hardware" class="md-nav__link">
    <span class="md-ellipsis">
      Optimizing GEMM for a given hardware
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#run-openblas-test" class="md-nav__link">
    <span class="md-ellipsis">
      Run OpenBLAS Test
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#benchmarking" class="md-nav__link">
    <span class="md-ellipsis">
      Benchmarking
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#adding-autodetection-support-for-a-new-revision-or-variant-of-a-supported-cpu" class="md-nav__link">
    <span class="md-ellipsis">
      Adding autodetection support for a new revision or variant of a supported cpu
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#adding-dedicated-support-for-a-new-cpu-model" class="md-nav__link">
    <span class="md-ellipsis">
      Adding dedicated support for a new cpu model
    </span>
  </a>
  
</li>
      
        <li class="md-nav__item">
  <a href="#adding-support-for-an-entirely-new-architecture" class="md-nav__link">
    <span class="md-ellipsis">
      Adding support for an entirely new architecture
    </span>
  </a>
  
</li>
      
    </ul>
  
</nav>
                  </div>
                </div>
              </div>
            
          
            <div class="md-content" data-md-component="content">
              <article class="md-content__inner md-typeset">
                
                  
<h1 id="developer-manual">Developer manual</h1>
<h2 id="source-codes-layout">Source codes Layout</h2>
<div class="highlight"><pre><span></span><code>OpenBLAS/  
├── benchmark                  Benchmark codes for BLAS
├── cmake                      CMakefiles
├── ctest                      Test codes for CBLAS interfaces
├── driver                     Implemented in C
│   ├── level2
│   ├── level3
│   ├── mapper
│   └── others                 Memory management, threading, etc
├── exports                    Generate shared library
├── interface                  Implement BLAS and CBLAS interfaces (calling driver or kernel)
│   ├── lapack
│   └── netlib
├── kernel                     Optimized assembly kernels for CPU architectures
│   ├── alpha                  Original GotoBLAS kernels for DEC Alpha
│   ├── arm                    ARMV5,V6,V7 kernels (including generic C codes used by other architectures)
│   ├── arm64                  ARMV8
│   ├── generic                General kernel codes written in plain C, parts used by many architectures.
│   ├── ia64                   Original GotoBLAS kernels for Intel Itanium
│   ├── mips
│   ├── mips64
│   ├── power
|   ├── riscv64
|   ├── simd                   Common code for Universal Intrinsics, used by some x86_64 and arm64 kernels
│   ├── sparc
│   ├── x86
│   ├── x86_64
│   └── zarch   
├── lapack                      Optimized LAPACK codes (replacing those in regular LAPACK)
│   ├── getf2
│   ├── getrf
│   ├── getrs
│   ├── laswp
│   ├── lauu2
│   ├── lauum
│   ├── potf2
│   ├── potrf
│   ├── trti2
│   ├── trtri
│   └── trtrs
├── lapack-netlib               LAPACK codes from netlib reference implementation
├── reference                   BLAS Fortran reference implementation (unused)
├── relapack                    Elmar Peise&#39;s recursive LAPACK (implemented on top of regular LAPACK)
├── test                        Test codes for BLAS
└── utest                       Regression test
</code></pre></div>
<p>A call tree for <code>dgemm</code> is as following.</p>
<div class="highlight"><pre><span></span><code>interface/gemm.c
        │
driver/level3/level3.c
        │
gemm assembly kernels at kernel/
</code></pre></div>
<p>To find the kernel currently used for a particular supported cpu, please check the corresponding <code>kernel/$(ARCH)/KERNEL.$(CPU)</code> file.</p>
<p>Here is an example for <code>kernel/x86_64/KERNEL.HASWELL</code></p>
<p><div class="highlight"><pre><span></span><code>...
DTRMMKERNEL    =  dtrmm_kernel_4x8_haswell.c
DGEMMKERNEL    =  dgemm_kernel_4x8_haswell.S
...
</code></pre></div>
According to the above <code>KERNEL.HASWELL</code>, OpenBLAS Haswell dgemm kernel file is <code>dgemm_kernel_4x8_haswell.S</code>.</p>
<h2 id="optimizing-gemm-for-a-given-hardware">Optimizing GEMM for a given hardware</h2>
<p>Read the Goto paper to understand the algorithm.</p>
<p>Goto, Kazushige; van de Geijn, Robert A. (2008). <a href="http://delivery.acm.org/10.1145/1360000/1356053/a12-goto.pdf?ip=155.68.162.54&amp;id=1356053&amp;acc=ACTIVE%20SERVICE&amp;key=A79D83B43E50B5B8%2EF070BBE7E45C3F17%2E4D4702B0C3E38B35%2E4D4702B0C3E38B35&amp;__acm__=1517932837_edfe766f1e295d9a7830812371e1d173">"Anatomy of High-Performance Matrix Multiplication"</a>. ACM Transactions on Mathematical Software 34 (3): Article 12
(The above link is available only to ACM members, but this and many related papers is also available on the pages
of van de Geijn's FLAME project, http://www.cs.utexas.edu/~flame/web/FLAMEPublications.html )</p>
<p>The <code>driver/level3/level3.c</code> is the implementation of Goto's algorithm. Meanwhile, you can look at <code>kernel/generic/gemmkernel_2x2.c</code>, which is a naive <code>2x2</code> register blocking gemm kernel in C.</p>
<p>Then,
* Write optimized assembly kernels. consider instruction pipeline, available registers, memory/cache accessing
* Tuning cache block size, <code>Mc</code>, <code>Kc</code>, and <code>Nc</code> </p>
<p>Note that not all of the cpu-specific parameters in param.h are actively used in algorithms. DNUMOPT only appears as a scale factor in profiling output of the level3 syrk interface code, while its counterpart SNUMOPT (aliased as NUMOPT in common.h) is not used anywhere at all. 
SYMV_P is only used in the generic kernels for the symv and chemv/zhemv functions - at least some of those are usually overridden by cpu-specific implementations, so if you start by cloning the existing implementation for a related cpu you need to check its KERNEL file to see if tuning SYMV_P would have any effect at all. 
GEMV_UNROLL is only used by some older x86_64 kernels, so not all sections in param.h define it.
Similarly, not all of the cpu parameters like L2 or L3 cache sizes are necessarily used in current kernels for a given model - by all indications the cpu identification code was imported from some other project originally.</p>
<h2 id="run-openblas-test">Run OpenBLAS Test</h2>
<p>We use netlib blas test, cblas test, and LAPACK test. Meanwhile, we use <a href="https://github.com/xianyi/BLAS-Tester">BLAS-Tester</a>, a modified test tool from ATLAS.</p>
<ul>
<li>Run <code>test</code> and <code>ctest</code> at OpenBLAS. e.g. <code>make test</code> or <code>make ctest</code>.</li>
<li>Run regression test <code>utest</code> at OpenBLAS.</li>
<li>Run LAPACK test. e.g. <code>make lapack-test</code>.</li>
<li>Clone <a href="https://github.com/xianyi/BLAS-Tester">BLAS-Tester</a>, which can compare the OpenBLAS result with netlib reference BLAS.</li>
</ul>
<p>The project makes use of several Continuous Integration (CI) services conveniently interfaced with github to automatically check compilability on a number of platforms.
Lastly, the testsuites included with "numerically heavy" projects like Julia, NumPy, Octave or QuantumEspresso can be used for regression testing.</p>
<h2 id="benchmarking">Benchmarking</h2>
<p>Several simple C benchmarks for performance testing individual BLAS functions are available in the <code>benchmark</code> folder, and its <code>scripts</code> subdirectory contains corresponding versions for Python, Octave and R.
Other options include</p>
<ul>
<li>https://github.com/RoyiAvital/MatlabJuliaMatrixOperationsBenchmark (various matrix operations in Julia and Matlab)</li>
<li>https://github.com/mmperf/mmperf/ (single-core matrix multiplication)</li>
</ul>
<h2 id="adding-autodetection-support-for-a-new-revision-or-variant-of-a-supported-cpu">Adding autodetection support for a new revision or variant of a supported cpu</h2>
<p>Especially relevant for x86_64, a new cpu model may be a "refresh" (die shrink and/or different number of cores) within an existing
model family without significant changes to its instruction set. (e.g. Intel Skylake, Kaby Lake etc. still are fundamentally Haswell,
low end Goldmont etc. are Nehalem). In this case, compilation with the appropriate older TARGET will already lead to a satisfactory build.</p>
<p>To achieve autodetection of the new model, its CPUID (or an equivalent identifier) needs to be added in the <code>cpuid_&lt;architecture&gt;.c</code>
relevant for its general architecture, with the returned name for the new type set appropriately. For x86 which has the most complex
cpuid file, there are two functions that need to be edited - get_cpuname() to return e.g. CPUTYPE_HASWELL and get_corename() for the (broader)
core family returning e.g. CORE_HASWELL. (This information ends up in the Makefile.conf and config.h files generated by <code>getarch</code>. Failure to
set either will typically lead to a missing definition of the GEMM_UNROLL parameters later in the build, as <code>getarch_2nd</code> will be unable to
find a matching parameter section in param.h.)</p>
<p>For architectures where "DYNAMIC_ARCH" builds are supported, a similar but simpler code section for the corresponding runtime detection of the cpu exists in <code>driver/others/dynamic.c</code> (for x86) and <code>driver/others/dynamic_&lt;arch&gt;.c</code> for other architectures.<br />
Note that for x86 the CPUID is compared after splitting it into its family, extended family, model and extended model parts, so the single decimal
number returned by Linux in /proc/cpuinfo for the model has to be converted back to hexadecimal before splitting into its constituent
digits, e.g. 142 = 8E , translates to extended model 8, model 14.</p>
<h2 id="adding-dedicated-support-for-a-new-cpu-model">Adding dedicated support for a new cpu model</h2>
<p>Usually it will be possible to start from an existing model, clone its KERNEL configuration file to the new name to use for this TARGET and eventually replace individual kernels with versions better suited for peculiarities of the new cpu model. In addition, it is necessary to add
(or clone at first) the corresponding section of GEMM_UNROLL parameters in the toplevel param.h, and possibly to add definitions such as USE_TRMM
(governing whether TRMM functions use the respective GEMM kernel or a separate source file) to the Makefiles (and CMakeLists.txt) in the kernel
directory. The new cpu name needs to be added to TargetLists.txt and the cpu autodetection code used by the <code>getarch</code> helper program - contained in
the <code>cpuid_&lt;architecture&gt;.c</code> file amended to include the CPUID (or equivalent) information processing required (see preceding section).</p>
<h2 id="adding-support-for-an-entirely-new-architecture">Adding support for an entirely new architecture</h2>
<p>This endeavour is best started by cloning the entire support structure for 32bit ARM, and within that the ARMV5 cpu in particular as this is implemented through plain C kernels only. An example providing a convenient "shopping list" can be seen in pull request #1526.</p>


              </article>
            </div>
          
          
<script>var target=document.getElementById(location.hash.slice(1));target&&target.name&&(target.checked=target.name.startsWith("__tabbed_"))</script>
        </div>
        
      </main>
      
        <footer class="md-footer">
  
  <div class="md-footer-meta md-typeset">
    <div class="md-footer-meta__inner md-grid">
      <div class="md-copyright">
  
  
    Made with
    <a href="https://squidfunk.github.io/mkdocs-material/" target="_blank" rel="noopener">
      Material for MkDocs
    </a>
  
</div>
      
    </div>
  </div>
</footer>
      
    </div>
    <div class="md-dialog" data-md-component="dialog">
      <div class="md-dialog__inner md-typeset"></div>
    </div>
    
    
    <script id="__config" type="application/json">{"base": "..", "features": [], "search": "../assets/javascripts/workers/search.b8dbb3d2.min.js", "translations": {"clipboard.copied": "Copied to clipboard", "clipboard.copy": "Copy to clipboard", "search.result.more.one": "1 more on this page", "search.result.more.other": "# more on this page", "search.result.none": "No matching documents", "search.result.one": "1 matching document", "search.result.other": "# matching documents", "search.result.placeholder": "Type to start searching", "search.result.term.missing": "Missing", "select.version": "Select version"}}</script>
    
    
      <script src="../assets/javascripts/bundle.5cfa9459.min.js"></script>
      
    
  </body>
</html>