<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    +1 to what Danny said<br>
    <br>
    <div class="moz-cite-prefix">On 04/19/2016 06:49 AM, Daniel Berlin
      via llvm-dev wrote:<br>
    </div>
    <blockquote
cite="mid:CAF4BwTXpZAz+is+oHRSaihcz9gEOj0Ja_oK68BeGy0LxfpAs3g@mail.gmail.com"
      type="cite">
      <div dir="ltr">Necessary super-linear behavior in analysis is
        pretty rare.
        <div>When it does occur, it should mostly be occurring on
          strange and wonderful CFGs.</div>
        <div><br>
        </div>
        <div>IE andersen's is N^3. You are highly unlikely to find *any*
          real world cases that trigger N^3 behavior out of the
          andersen's implementation in GCC.</div>
        <div><br>
          <div>Honestly, at the point where you've added more than 1-2
            knobs to something, it's time to reconsider whether that's
            the best way to do it anymore.<br>
          </div>
          <div><br>
          </div>
        </div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Tue, Apr 19, 2016 at 6:19 AM, Joerg
          Sonnenberger via llvm-dev <span dir="ltr"><<a
              moz-do-not-send="true"
              href="mailto:llvm-dev@lists.llvm.org" target="_blank"><a class="moz-txt-link-abbreviated" href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a></a>></span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex"><span
              class="">On Mon, Apr 18, 2016 at 04:37:58PM -0700, Philip
              Reames wrote:<br>
              > Cases like this illustrate why we should avoid
              arbitrary cutoffs in analysis<br>
              > passes if at all possible.<br>
              <br>
            </span>I fully agree on that, but it doesn't help if
            super-linear behavior,<br>
            both in terms of CPU time and memory use practically happen.
            It doesn't<br>
            mean that we couldn't e.g. tune the limits based on the
            optimisation<br>
            level, but currently it is kind of all or nothing.<br>
            <div class="HOEnZb">
              <div class="h5"><br>
                Joerg<br>
                _______________________________________________<br>
                LLVM Developers mailing list<br>
                <a moz-do-not-send="true"
                  href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a><br>
                <a moz-do-not-send="true"
                  href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev"
                  rel="noreferrer" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
              </div>
            </div>
          </blockquote>
        </div>
        <br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
LLVM Developers mailing list
<a class="moz-txt-link-abbreviated" href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a>
<a class="moz-txt-link-freetext" href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>