<html>
  <head>
    <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    On 2012-01-20 17:46, Bengt Rutisson wrote:
    <blockquote cite="mid:4F199A60.6070506@oracle.com" type="cite">
      <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
      <br>
      Stefan,<br>
      <br>
      Thanks for the prompt review!<br>
      <br>
      Comments inline.<br>
      <br>
      On 2012-01-20 17:21, Stefan Karlsson wrote:
      <blockquote
        cite="mid:8C7CE33B-1215-4828-836B-EC0860F5D5CF@oracle.com"
        type="cite">
        <div><span class="Apple-style-span"
            style="-webkit-tap-highlight-color: rgba(26, 26, 26,
            0.296875); -webkit-composition-fill-color: rgba(175, 192,
            227, 0.230469); -webkit-composition-frame-color: rgba(77,
            128, 180, 0.230469); ">On 20 jan 2012, at 17:06, Bengt
            Rutisson <<a moz-do-not-send="true"
              href="mailto:bengt.rutisson@oracle.com">bengt.rutisson@oracle.com</a>>

            wrote:</span><br>
        </div>
        <div><br>
        </div>
        <blockquote type="cite">
          <div><span></span><br>
            <span>Hi all,</span><br>
            <span></span><br>
            <span>Can I have a couple of quick reviews for this small
              change:</span><br>
            <span><a moz-do-not-send="true"
                href="http://cr.openjdk.java.net/%7Ebrutisso/7131791/webrev.02">http://cr.openjdk.java.net/~brutisso/7131791/webrev.02</a></span><br>
            <span></span><br>
            <span>This should hopefully fix the 500+ failures in the G1
              nightlies. So, I would like to get it in before the
              nightlies tonight.</span><br>
            <span></span><br>
            <span>The issue is that we call collect() which will trigger
              a collection without protecting the memory that we just
              allocated for a humongous object. The fix (thanks Tony for
              helping me out!!!) is to fake an object and create a
              handle to it before we call collect.</span><br>
          </div>
        </blockquote>
        <div><br>
        </div>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;"><span class="changed">1067         Handle h((oop)result);</span> </span></font></pre>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">1068         collect(GCCause::_g1_humongous_allocation); </span></font></pre>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">1069       } </span></font></pre>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">1070       return result; </span></font></pre>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">1071     }</span></font></pre>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">Can we really have a handle to uninitialized memory?</span></font></pre>
      </blockquote>
      <br>
      The memory is not uninitialized since I fake an object there with
      the call to CollectedHeap::fill_with_object(result, word_size,
      false); just before the code you have above.<br>
    </blockquote>
    <br>
    I missed that.<br>
    <br>
    <blockquote cite="mid:4F199A60.6070506@oracle.com" type="cite"> <br>
      <blockquote
        cite="mid:8C7CE33B-1215-4828-836B-EC0860F5D5CF@oracle.com"
        type="cite">
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">Are you sure that the humongous object will not be moved by a full collection. You should probably return h() instead of result.</span></font></pre>
      </blockquote>
      <br>
      Humongous objects will not be moved by G1 collections so I think
      we are ok.<br>
    </blockquote>
    <br>
    OK. But maybe we should be a bit defensive and return h() here.<br>
    <br>
    StefanK<br>
    <br>
    <blockquote cite="mid:4F199A60.6070506@oracle.com" type="cite"> <br>
      Thanks for the prompt review!<br>
      <br>
      Bengt<br>
      <blockquote
        cite="mid:8C7CE33B-1215-4828-836B-EC0860F5D5CF@oracle.com"
        type="cite">
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">
</span></font></pre>
        <pre><font class="Apple-style-span" face="Helvetica"><span class="Apple-style-span" style="white-space: normal;">StefanK</span></font></pre>
        <br>
        <blockquote type="cite">
          <div><span></span><br>
            <span>Bengt</span><br>
          </div>
        </blockquote>
      </blockquote>
      <br>
    </blockquote>
    <br>
  </body>
</html>