<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    Hello,<br>
    <br>
    This looks great! I'll add it to our tests during the weekend and if
    nothing pops up merge it next week.<br>
    <br>
    Lukas<br>
    <br>
    <div class="moz-cite-prefix">On 22/08/13 20:58, Steve Vinoski wrote:<br>
    </div>
    <blockquote
cite="mid:CAO+zUOVPn5V_MGWrH_+jEghf-FUfWpbdG-UrSXBF2jPGZzaNDA@mail.gmail.com"
      type="cite">
      <div dir="ltr">I've amended the commit on my branch as described
        in my previous message, please refetch.
        <div><br>
        </div>
        <div><a moz-do-not-send="true"
            href="https://github.com/erlang/otp/pull/58">https://github.com/erlang/otp/pull/58</a><br>
        </div>
        <div><br>
        </div>
        <div>thanks,</div>
        <div>--steve</div>
        <div><br>
        </div>
      </div>
      <div class="gmail_extra"><br>
        <br>
        <div class="gmail_quote">On Thu, Aug 22, 2013 at 12:55 PM, Steve
          Vinoski <span dir="ltr"><<a moz-do-not-send="true"
              href="mailto:vinoski@ieee.org" target="_blank">vinoski@ieee.org</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 Lukas,
              <div><br>
              </div>
              <div>Thanks for the feedback. After our IRC conversation
                today covering the details of this issue, I agree it's
                clearer if this patch doesn't introduce order
                dependencies between +S and +SP options or introduce
                accumulated effects of multiple +SP options. I'll change
                my branch to do the following:</div>
              <div><br>
              </div>
              <div>* ensure that later +SP options on the command line
                completely replace any previous +SP options, so there's
                no accumulated effects between them (which implies "+SP
                100:100" can undo any previous +SP options)</div>
              <div><br>
              </div>
              <div>* keep interactions between +S and +SP options, but
                remove ordering dependencies, thus your example "+S
                16:16 +SP 50:50 +S 16:16" would result in [8:8] as
                you've specified, as would both "+SP 50:50 +S 16:16" and
                "+S 16:16 +SP 50:50"</div>
              <div><br>
              </div>
              <div>* document and add tests for the already-existing
                feature that +S 0:0 undoes any prior +S options,
                resetting scheduler thread and scheduler thread online
                counts to their defaults for the host</div>
              <div>
                <br>
              </div>
              <div>* document and add tests for the already-existing
                feature that specifying negative numbers for +S results
                in the specified values being subtracted from the
                scheduler thread and scheduler thread online counts</div>
              <div><br>
              </div>
              <div>I'll send another email when the branch is ready with
                these changes. Any other concerns, please let me know.</div>
              <div><br>
              </div>
              <div>thanks,</div>
              <div>--steve</div>
              <div><br>
              </div>
            </div>
            <div class="HOEnZb">
              <div class="h5">
                <div class="gmail_extra">
                  <br>
                  <br>
                  <div class="gmail_quote">On Thu, Aug 22, 2013 at 10:11
                    AM, Lukas Larsson <span dir="ltr"><<a
                        moz-do-not-send="true"
                        href="mailto:lukas@erlang.org" target="_blank">lukas@erlang.org</a>></span>
                    wrote:<br>
                    <blockquote class="gmail_quote" style="margin:0 0 0
                      .8ex;border-left:1px #ccc solid;padding-left:1ex">
                      <div text="#000000" bgcolor="#FFFFFF"> Hello
                        Steve,<br>
                        <br>
                        Have you given any thoughts on what should
                        happen if you chain multiple +SP commands? i.e.<br>
                        <br>
                        erl +S 16:16 +SP 50:50 +SP 50:50<br>
                        <br>
                        Our view right now is that this should give
                        [8:8] and not [4:4] as it currently does in the
                        patch. <br>
                        <br>
                        Related to this we also do not think that the
                        order to +S vs +SP commands should matter. i.e.<br>
                        <br>
                        erl +S 16:16 +SP 50:50 +S 16:16<br>
                        <br>
                        should give [8:8] and not [16:16].<br>
                        <br>
                        What do you think? These things are for sure odd
                        cases to think about, but if it is possible
                        someone will for sure do it....<br>
                        <br>
                        Lukas
                        <div>
                          <div><br>
                            <br>
                            <div>On 21/08/13 15:45, Steve Vinoski wrote:<br>
                            </div>
                          </div>
                        </div>
                        <blockquote type="cite">
                          <div>
                            <div>
                              <div dir="ltr">
                                <div>
                                  <div>For applications where
                                    measurements show enhanced
                                    performance from the use of a
                                    non-default number of emulator
                                    scheduler threads, having to
                                    accurately set the right number of
                                    scheduler threads across multiple
                                    hosts each with different numbers of
                                    logical processors is difficult
                                    because the erl +S option requires
                                    absolute numbers of scheduler
                                    threads and scheduler threads online
                                    to be specified.</div>
                                  <div><br>
                                  </div>
                                  <div>To address this issue, this patch
                                    adds a +SP option to erl, similar to
                                    the existing +S option but allowing
                                    the number of scheduler threads and
                                    scheduler threads online to be set
                                    as percentages of logical processors
                                    configured and logical processors
                                    available, respectively. For
                                    example, "+SP 50:25" sets the number
                                    of scheduler threads to 50% of the
                                    logical processors configured, and
                                    the number of scheduler threads
                                    online to 25% of the logical
                                    processors available.</div>
                                </div>
                                <div><br>
                                </div>
                                <div>
                                  <div><a moz-do-not-send="true"
                                      href="https://github.com/erlang/otp/pull/58"
                                      target="_blank">https://github.com/erlang/otp/pull/58</a></div>
                                  <div><br>
                                  </div>
                                </div>
                                <div>--steve</div>
                              </div>
                              <br>
                              <fieldset></fieldset>
                              <br>
                            </div>
                          </div>
                          <pre>_______________________________________________
erlang-patches mailing list
<a moz-do-not-send="true" href="mailto:erlang-patches@erlang.org" target="_blank">erlang-patches@erlang.org</a>
<a moz-do-not-send="true" href="http://erlang.org/mailman/listinfo/erlang-patches" target="_blank">http://erlang.org/mailman/listinfo/erlang-patches</a>
</pre>
                        </blockquote>
                        <br>
                      </div>
                    </blockquote>
                  </div>
                  <br>
                </div>
              </div>
            </div>
          </blockquote>
        </div>
        <br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
erlang-patches mailing list
<a class="moz-txt-link-abbreviated" href="mailto:erlang-patches@erlang.org">erlang-patches@erlang.org</a>
<a class="moz-txt-link-freetext" href="http://erlang.org/mailman/listinfo/erlang-patches">http://erlang.org/mailman/listinfo/erlang-patches</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>