<div dir="ltr">Tao, <div><br></div><div>CMS no longer use InitialTenuringThreshold.</div><div>IIRC, even before when CMS used InitialTenuringThreshold, setting initial and max the same value didn't turn off the adaptiveness.</div><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Mar 10, 2015 at 12:23 AM, Tao Mao <span dir="ltr"><<a href="mailto:yiyeguhu@gmail.com" target="_blank">yiyeguhu@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi Jungwoo,<div><br></div><div>Would fixing InitialTenuringThreshold and MaxTenuringThreshold to the same value work for you?</div><div><br></div><div>Thanks.</div><span class="HOEnZb"><font color="#888888"><div>Tao Mao</div><div><br></div></font></span></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Mar 9, 2015 at 10:39 PM, Jungwoo Ha <span dir="ltr"><<a href="mailto:jwha@google.com" target="_blank">jwha@google.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">We were investigating some features and figured out that ParallelScavange and ParNew has different TenuringTreshold heuristics.<div>We thought it would be nice to have a flag that can control the adaptiveness, so that we can fix it on purpose to isolate other performance factors.</div><span><font color="#888888"><div>--Jungwoo</div></font></span></div><div><div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Mar 9, 2015 at 11:27 AM, Jon Masamitsu <span dir="ltr"><<a href="mailto:jon.masamitsu@oracle.com" target="_blank">jon.masamitsu@oracle.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
  
    
  
  <div bgcolor="#FFFFFF" text="#000000">
    Jungwoo,<span><br>
    <br>
    <div>On 03/06/2015 06:03 PM, Jungwoo Ha
      wrote:<br>
    </div>
    <blockquote type="cite">
      <div dir="ltr">I think it is the same for all JDK{7,8,9}.
        <div><br>
        </div>
        <div>arguments.cpp disabled UseAdaptiveSizePolicy for CMS.</div>
        <div>parallel scavange initializes _tenuring_threshold with</div>
        <div>  _tenuring_threshold = (UseAdaptiveSizePolicy) ?
          InitialTenuringThreshold : MaxTenuringThreshold;</div>
        <div>whereas CMS uses</div>
        <div>  _tenuring_threshold = MaxTenuringThreshold;</div>
        <div><br>
        </div>
        <div>adjust_desired_tenuring_threshold() is called inside
          ParNewGeneration::collect(), which calls
          compute_tenuring_threshold.</div>
        <div>It only uses MaxTenuringThreshold and TargetSurvivorRatio.</div>
        <div>I don't see any other flags here.<br>
        </div>
      </div>
    </blockquote>
    <br></span>
    Yes, this is the intended behavior.<br>
    <br>
    You're saying we're sometimes wasting survivor space?<span><font color="#888888"><br>
    <br>
    Jon</font></span><span><br>
    <br>
    <blockquote type="cite">
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Fri, Mar 6, 2015 at 4:28 PM, Jon
          Masamitsu <span dir="ltr"><<a href="mailto:jon.masamitsu@oracle.com" target="_blank">jon.masamitsu@oracle.com</a>></span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">This is in
            jdk9?<span><font color="#888888"><br>
                <br>
                Jon</font></span>
            <div>
              <div><br>
                <br>
                On 3/6/2015 4:21 PM, Jungwoo Ha wrote:<br>
                <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
                  On CMS when UseAdaptiveSizePolicy is off, still
                  tenuring_threshold is being changed.<br>
                  Is this an intended behavior?<br>
                  (yes. I know it is not "size" related heuristics)<br>
                  <br>
                  It would be nice to have a flag that can disable
                  adjusting tenuring threshold,<br>
                  or at least set minimum threshold. When the default
                  heuristics doesn't work,<br>
                  there are nothing much we can do other than changing
                  the Max.<br>
                  <br>
                  --Jungwoo<br>
                </blockquote>
                <br>
              </div>
            </div>
          </blockquote>
        </div>
        <br>
      </div>
    </blockquote>
    <br>
  </span></div>

</blockquote></div><br></div>
</div></div></blockquote></div><br></div>
</div></div></blockquote></div><br></div>