<html>
    <head>
      <base href="http://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 --- - pcsa.f and csa.f from xplor-nih compile ~100x too slowly using -fplugin-arg-dragonegg-llvm-option=-vectorize"
   href="http://llvm.org/bugs/show_bug.cgi?id=15222">15222</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>pcsa.f and csa.f from xplor-nih compile ~100x too slowly using -fplugin-arg-dragonegg-llvm-option=-vectorize
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>dragonegg
          </td>
        </tr>

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

        <tr>
          <th>Hardware</th>
          <td>Macintosh
          </td>
        </tr>

        <tr>
          <th>OS</th>
          <td>MacOS X
          </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>New Bugs
          </td>
        </tr>

        <tr>
          <th>Assignee</th>
          <td>baldrick@free.fr
          </td>
        </tr>

        <tr>
          <th>Reporter</th>
          <td>howarth@nitro.med.uc.edu
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvmbugs@cs.uiuc.edu
          </td>
        </tr>

        <tr>
          <th>Classification</th>
          <td>Unclassified
          </td>
        </tr></table>
      <p>
        <div>
        <pre>Created <span class=""><a href="attachment.cgi?id=9989" name="attach_9989" title="test cases exposing the vectorization bottleneck">attachment 9989</a> <a href="attachment.cgi?id=9989&action=edit" title="test cases exposing the vectorization bottleneck">[details]</a></span>
test cases exposing the vectorization bottleneck

The attached code from xplor-nih takes an abnormally long time to compile with
-fplugin-arg-dragonegg-llvm-option=-vectorize under dragonegg...

% time de-gfortran47 -c pcsa.f -Ofast
-fplugin-arg-dragonegg-llvm-option=-vectorize -fdefault-integer-8
-ffixed-line-length-120 -fno-second-underscore
80.241u 0.237s 1:20.49 99.9%    0+0k 0+22io 0pf+0w

% time de-gfortran47 -c csa.f -Ofast
-fplugin-arg-dragonegg-llvm-option=-vectorize -fdefault-integer-8
-ffixed-line-length-120 -fno-second-underscore
285.138u 0.573s 4:45.73 99.9%   0+0k 0+13io 0pf+0w

compared to the normal compilation without vectorization...

% time de-gfortran47 -c pcsa.f -Ofast -fdefault-integer-8
-ffixed-line-length-120 -fno-second-underscore
0.560u 0.043s 0:00.60 100.0%    0+0k 0+20io 0pf+0w

% time de-gfortran47 -c csa.f -Ofast -fdefault-integer-8
-ffixed-line-length-120 -fno-second-underscore
0.842u 0.059s 0:00.90 98.8%     0+0k 0+10io 0pf+0w

These may be good test cases for finding bottlenecks in the current vectorizor
code. Note that stock gfortran 4.7.2 compiles this same code with vectorization
very rapidly...

% time gfortran-fsf-4.7 -c pcsa.f -Ofast -fdefault-integer-8
-ffixed-line-length-120 -fno-second-underscore
1.187u 0.049s 0:01.29 94.5%    0+0k 1+20io 528pf+0w

% time gfortran-fsf-4.7 -c csa.f -Ofast -fdefault-integer-8
-ffixed-line-length-120 -fno-second-underscore
2.459u 0.083s 0:02.54 99.6%    0+0k 0+8io 0pf+0w

Tested on x86_64-apple-darwin12 with llvm/dragonegg svn at r174816.</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>