<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    On 2012-05-08 07:28, Kirk Pepperdine wrote:
    <blockquote
      cite="mid:B51B6C47-9BF5-49F9-B9EF-A9F58F7ACA70@kodewerk.com"
      type="cite">This <a moz-do-not-send="true"
        href="http://monaco.us.oracle.com/detail.jsp?cr=7166894"
        style="font-family: Times; font-size: 13px; font-style: normal;
        font-variant: normal; font-weight: normal; letter-spacing:
        normal; line-height: normal; orphans: 2; text-align:
        -webkit-auto; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px;
        background-color: rgb(238, 238, 238); ">7166894: Add gc cause to
        Full GC logging for all collectors</a> leads me to an internal
      Oracle server.</blockquote>
    <br>
    This is unfortunate, but it is a consequence of the fact that we
    still do not have a publicly available JIRA installation for OpenJDK
    issues. As you are aware there is active work on creating such an
    installation.<br>
    <br>
    In the mean time the public access to the bug system is through
    bugs.sun.com, which is lagging several days behind. The current bug
    is still not available there, but will hopefully be soon, on this
    URL:<br>
    <br>
    <a class="moz-txt-link-freetext" href="http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7166894">http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7166894</a><br>
    <br>
    Bengt<br>
    <br>
    <blockquote
      cite="mid:B51B6C47-9BF5-49F9-B9EF-A9F58F7ACA70@kodewerk.com"
      type="cite">
      <div><br>
      </div>
      <div>Regards,</div>
      <div>Kirk</div>
      <div><br>
        <div>
          <div>On 2012-05-07, at 10:09 PM, Bengt Rutisson wrote:</div>
          <br class="Apple-interchange-newline">
          <blockquote type="cite">
            <meta content="text/html; charset=ISO-8859-1"
              http-equiv="Content-Type">
            <div bgcolor="#FFFFFF" text="#000000"> <br>
              Hi Vitaly and Mikael,<br>
              <br>
              Thanks both of you for looking at this! Here is an updated
              webrev using jio_snprintf:<br>
              <a moz-do-not-send="true" class="moz-txt-link-freetext"
                href="http://cr.openjdk.java.net/%7Ebrutisso/7166894/webrev.01/">http://cr.openjdk.java.net/~brutisso/7166894/webrev.01/</a><br>
              <br>
              On 2012-05-07 15:45, Vitaly Davidovich wrote:
              <blockquote
cite="mid:CAHjP37Fd=7q5we=_f7hSuQeefKD-sUWP5muP3cf=kYTwzY+B4g@mail.gmail.com"
                type="cite">
                <p>Yup, I don't think it would simplify the code but
                  rather make any potential maintenance easier, I
                  suspect.  For example (and it's a stretch), if you
                  wanted to change the size of the message buffer, you'd
                  modify just one place.  However, this is very minor so
                  your call.</p>
              </blockquote>
              <br>
              When I went back to look at the G1 code I think that code
              will benefit even more from some abstraction like you
              suggested. But on the other hand there is a lot of logging
              code that would benefit from that. So, I have to think a
              bit about how much I would like to change. I'll leave it
              as it is for now.<br>
              <br>
              Thanks again,<br>
              Bengt<br>
              <br>
              <blockquote
cite="mid:CAHjP37Fd=7q5we=_f7hSuQeefKD-sUWP5muP3cf=kYTwzY+B4g@mail.gmail.com"
                type="cite">
                <p>Thanks,</p>
                <p>Vitaly</p>
                <p>Sent from my phone</p>
                <div class="gmail_quote">On May 7, 2012 9:41 AM, "Bengt
                  Rutisson" <<a moz-do-not-send="true"
                    href="mailto:bengt.rutisson@oracle.com">bengt.rutisson@oracle.com</a>>

                  wrote:<br type="attribution">
                  <blockquote class="gmail_quote" style="margin:0 0 0
                    .8ex;border-left:1px #ccc solid;padding-left:1ex">
                    <div bgcolor="#ffffff" text="#000000"> <br>
                      Hi Vitaly,<br>
                      <br>
                      Thanks for looking at this!<br>
                      <br>
                      On 2012-05-07 15:31, Vitaly Davidovich wrote:
                      <blockquote type="cite">
                        <p>Hi Bengt,</p>
                        <p>The allocation of the buffer to hold the
                          string and sprintf'ing to it is repeated at
                          the places you changed.  Do you think it's
                          worth it to factor that out into a common
                          function or maybe some stack allocated class
                          whose destructor would free the message
                          buffer? Just a thought ...</p>
                      </blockquote>
                      <br>
                      I see your point, but I'm not sure this would
                      actually simplify the code.<br>
                      <br>
                      <blockquote type="cite">
                        <p>Also, probably better to use snprintf instead
                          of sprintf.</p>
                      </blockquote>
                      <br>
                      Good point. I'll fix this. This code fragment is
                      actually copied from existing G1 code, so I'll go
                      back and change that too.<br>
                      <br>
                      Thanks again for looking at it!<br>
                      Bengt<br>
                      <br>
                      <blockquote type="cite">
                        <p>Regards,</p>
                        <p>Vitaly</p>
                        <p>Sent from my phone</p>
                        <div class="gmail_quote">On May 7, 2012 8:35 AM,
                          "Bengt Rutisson" <<a moz-do-not-send="true"
                            href="mailto:bengt.rutisson@oracle.com"
                            target="_blank">bengt.rutisson@oracle.com</a>>

                          wrote:<br type="attribution">
                          <blockquote class="gmail_quote"
                            style="margin:0pt 0pt 0pt
                            0.8ex;border-left:1px solid
                            rgb(204,204,204);padding-left:1ex"> <br>
                            Hi all,<br>
                            <br>
                            Can I get a couple of reviews of this simple
                            change:<br>
                            <a moz-do-not-send="true"
                              href="http://cr.openjdk.java.net/%7Ebrutisso/7166894/webrev/"
                              target="_blank">http://cr.openjdk.java.net/~brutisso/7166894/webrev/</a><br>
                            <br>
                            Background:<br>
                            I recently pushed a similar fix for G1 as
                            "7163848: G1: Log GC Cause for a GC". That
                            fix adds the GC cause information to all G1
                            GCs. It was discussed if we should do this
                            for all collectors and we came to the
                            conclusion that it would be fairly safe to
                            do it for all Full GCs. These log messages
                            already contain the text "(System)" or
                            "(System.gc())" when a System.gc() happens
                            and PrintGCDetails are enabled. Now they
                            will always contain the cause in
                            parenthesis, even when only PrintGC is
                            enabled. Hopefully most parsing will work
                            with this.<br>
                            <br>
                            Thanks,<br>
                            Bengt<br>
                          </blockquote>
                        </div>
                      </blockquote>
                      <br>
                    </div>
                  </blockquote>
                </div>
              </blockquote>
              <br>
            </div>
          </blockquote>
        </div>
        <br>
      </div>
    </blockquote>
    <br>
  </body>
</html>