<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 - Loop with early continue does not produce same IR as loop without"
   href="https://bugs.llvm.org/show_bug.cgi?id=45255">45255</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>Loop with early continue does not produce same IR as loop without
          </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>All
          </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>Loop Optimizer
          </td>
        </tr>

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

        <tr>
          <th>Reporter</th>
          <td>Matthew.Arsenault@amd.com
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvm-bugs@lists.llvm.org
          </td>
        </tr></table>
      <p>
        <div>
        <pre>Created <span class=""><a href="attachment.cgi?id=23260" name="attach_23260" title="Testcase with early continue">attachment 23260</a> <a href="attachment.cgi?id=23260&action=edit" title="Testcase with early continue">[details]</a></span>
Testcase with early continue

This testcase has two versions of a program with a slightly different loop
structure.

The first testcase (extract.while.ll) looks like this:

while (I < N) {
  if (foo()) {
    ++I;
    continue;
  }

  bar();
  ++I;
}

The second version (extract.while.no.continue.ll looks like this:

while (I < N) {
  if (!foo()) {
    bar();
  }

  ++I;
}

I would expect these to optimize to the same IR, but they do not. The form with
the continue ends up codegenning to something worse, and ends up about 50%
slower on AMDGPU.

Note the attached IR has already gone through the optimizer once, and the
second form was rotated and has the preheader. I've been operating under the
assumption I should be able to just run -loop-rotate on the first form to get
equivalent IR as the second form.


The relevant part of the testcase is while.body.i/if.then320.i. if.then320.i is
the continue block in the first version.

I'm able to get opt -loop-rotate to rotate this loop by increasing
-rotation-max-header-size to at least 18, and hacking
profitableToRotateLoopExitingLatch to always return true</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>