<html>
    <head>
      <base href="https://llvm.org/bugs/" />
    </head>
    <body><table border="1" cellspacing="0" cellpadding="8">
        <tr>
          <th>Bug ID</th>
          <td><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW --- - [linux] clang should define _FP_FAST_FMA, _FP_FAST_FMAF, _FP_FAST_FMAL when fma(), fmaf(), fmal() is faster than an add and a mul"
   href="https://llvm.org/bugs/show_bug.cgi?id=25728">25728</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>[linux] clang should define _FP_FAST_FMA, _FP_FAST_FMAF, _FP_FAST_FMAL when fma(), fmaf(), fmal() is faster than an add and a mul
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>clang
          </td>
        </tr>

        <tr>
          <th>Version</th>
          <td>unspecified
          </td>
        </tr>

        <tr>
          <th>Hardware</th>
          <td>PC
          </td>
        </tr>

        <tr>
          <th>OS</th>
          <td>Linux
          </td>
        </tr>

        <tr>
          <th>Status</th>
          <td>NEW
          </td>
        </tr>

        <tr>
          <th>Severity</th>
          <td>normal
          </td>
        </tr>

        <tr>
          <th>Priority</th>
          <td>P
          </td>
        </tr>

        <tr>
          <th>Component</th>
          <td>Frontend
          </td>
        </tr>

        <tr>
          <th>Assignee</th>
          <td>unassignedclangbugs@nondot.org
          </td>
        </tr>

        <tr>
          <th>Reporter</th>
          <td>nicolasweber@gmx.de
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvm-bugs@lists.llvm.org
          </td>
        </tr>

        <tr>
          <th>Classification</th>
          <td>Unclassified
          </td>
        </tr></table>
      <p>
        <div>
        <pre>Linux mathdef.h contains:

/* The GCC 4.6 compiler will define __FP_FAST_FMA{,F,L} if the fma{,f,l}
   builtins are supported.  */
# if __FP_FAST_FMA
#  define FP_FAST_FMA 1
# endif

# if __FP_FAST_FMAF
#  define FP_FAST_FMAF 1
# endif

# if __FP_FAST_FMAL
#  define FP_FAST_FMAL 1
# endif


The comment isn't quite right; it looks like gcc will define these when the
builtin is replaced with an efficient intrinsic.

thakis@thakis:~/src/chrome/src$ cat test.c
#include <math.h>

#ifdef FP_FAST_FMAF
#warning on
#else
#warning off
#endif
thakis@thakis:~/src/chrome/src$ gcc -c test.c 
test.c:6:2: warning: #warning off [-Wcpp]
 #warning off
  ^
thakis@thakis:~/src/chrome/src$ gcc -c test.c -march=core-avx2
test.c:4:2: warning: #warning on [-Wcpp]
 #warning on
  ^



With clang on the other hand it appears that this is never fast, even though
clang does generate vfmadd213sd just fine when calling fma():

thakis@thakis:~/src/chrome/src$
third_party/llvm-build/Release+Asserts/bin/clang -c test.c
test.c:6:2: warning: off [-W#warnings]
#warning off
 ^
1 warning generated.
thakis@thakis:~/src/chrome/src$
third_party/llvm-build/Release+Asserts/bin/clang -c test.c -march=core-avx2
test.c:6:2: warning: off [-W#warnings]
#warning off
 ^
1 warning generated.</pre>
        </div>
      </p>
      <hr>
      <span>You are receiving this mail because:</span>
      
      <ul>
          <li>You are on the CC list for the bug.</li>
      </ul>
    </body>
</html>