<html>
    <head>
      <base href="https://bugs.llvm.org/">
    </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 - Vectorization improvement opportunity for max/min"
   href="https://bugs.llvm.org/show_bug.cgi?id=35284">35284</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>Vectorization improvement opportunity for max/min
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>libraries
          </td>
        </tr>

        <tr>
          <th>Version</th>
          <td>trunk
          </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>enhancement
          </td>
        </tr>

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

        <tr>
          <th>Component</th>
          <td>Backend: X86
          </td>
        </tr>

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

        <tr>
          <th>Reporter</th>
          <td>serguei.katkov@azul.com
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvm-bugs@lists.llvm.org
          </td>
        </tr></table>
      <p>
        <div>
        <pre>This is about possibly better vectorization coding for float/double max/min.

Let's consider the following c++ program (<a href="https://godbolt.org/g/yWxTCB">https://godbolt.org/g/yWxTCB</a>):
float test(float a[], int N) {
  float max = -1;
  for (int i = 0; i < N; i++)
    max = max < a[i] ? a[i] : max;
  return max;
}

clang trunk now generates a main loop with the following pattern:
movss + maxss (N times, where N might be 4 or 8).

icc 18 is able to use maxps packed instruction.

Even the code looking as:
movss 16 times (load values to xmm0-xmm15 registers)
ucommiss + ga (16 times)

and movaps in case of ga on slowpath

seems might be better.

I do not know what code would be best here but potentially there is an
opportunity here...</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>