<!DOCTYPE html>
<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <div class="moz-cite-prefix">On 2/4/26 10:26 AM, Ben Cooksley wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CA+XidOGtqOjws-+14LYwajipHpMBUCNSjSz8Y-2LELGi7yOuow@mail.gmail.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="ltr">
        <div dir="ltr">On Wed, Feb 4, 2026 at 9:07 PM Ben Cooksley <<a
            href="mailto:bcooksley@kde.org" moz-do-not-send="true"
            class="moz-txt-link-freetext">bcooksley@kde.org</a>>
          wrote:</div>
        <div class="gmail_quote gmail_quote_container">
          <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div dir="ltr">
              <div dir="ltr">On Wed, Feb 4, 2026 at 8:25 PM Vlad
                Zahorodnii <<a href="mailto:vlad.zahorodnii@kde.org"
                  target="_blank" moz-do-not-send="true"
                  class="moz-txt-link-freetext">vlad.zahorodnii@kde.org</a>>
                wrote:</div>
              <div class="gmail_quote">
                <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hello,<br>
                </blockquote>
                <div><br>
                </div>
                <div>HI Vlad,</div>
                <div> </div>
                <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                  <br>
                  Over approximately the past weekend something happened
                  in our CI and now <br>
                  it takes quite long time for tests to run. For
                  example, in kwin, we have <br>
                  a test that used to run for about 20 seconds, and now
                  it takes about 5 <br>
                  or so minutes to finish running. Speaking for kwin,
                  there were no <br>
                  changes that could increase test run times so
                  dramatically.<br>
                  <br>
                  January 26th:<br>
                  <br>
                           Start  61: kwin-testOutputChanges<br>
                    61/158 Test  #61: kwin-testOutputChanges <br>
                  .............................   Passed   19.36 sec<br>
                  <br>
                  January 29th:<br>
                  <br>
                           Start  61: kwin-testOutputChanges<br>
                    61/158 Test  #61: kwin-testOutputChanges <br>
                  .............................   Passed   43.93 sec<br>
                  <br>
                  January 30th:<br>
                  <br>
                           Start  61: kwin-testOutputChanges<br>
                    61/158 Test  #61: kwin-testOutputChanges <br>
                  .............................   Passed   45.91 sec<br>
                  <br>
                  Februrary 3rd:<br>
                  <br>
                           Start  61: kwin-testOutputChanges<br>
                    61/158 Test  #61: kwin-testOutputChanges <br>
                  .............................   Passed  254.19 sec<br>
                  <br>
                  FreeBSD appears to be fine.<br>
                  <br>
                  We suspect that test run times blew up due to enabling
                  LSAN in various <br>
                  libraries (kwin itself has no LSAN enabled yet). The
                  issue doesn't <br>
                  appear to be specific to only kwin, people reported
                  that they've seen <br>
                  similar issues in other projects too. Maybe something
                  else happened to <br>
                  CI that sysadmins will be able to clarify.<br>
                </blockquote>
                <div><br>
                </div>
                <div>Nothing else happened to CI recently aside from the
                  enablement of LSAN.</div>
                <div><br>
                </div>
                <div>The underlying SUSE images were for Qt 6.10 at
                  least last rebuilt on January 25th, which is well
                  before your "last good" date.</div>
                <div><br>
                </div>
                <div>The only change to CI between January 30th and
                  February 3rd was fast_unwind_on_malloc=0 being added
                  by default, even though it is primarily for the
                  benefit of LSAN.</div>
                <div>I've now made changes to only
                  set fast_unwind_on_malloc=0 if LSAN is explicitly
                  enabled for a repository - hard to tell if that will
                  fix the issue though as KWin takes a while to build.</div>
              </div>
            </div>
          </blockquote>
          <div><br>
          </div>
          <div>For the record, as per <a
              href="https://invent.kde.org/plasma/kwin/-/jobs/3958895"
              moz-do-not-send="true" class="moz-txt-link-freetext">https://invent.kde.org/plasma/kwin/-/jobs/3958895</a>
            which completed moments ago:</div>
        </div>
      </div>
    </blockquote>
    <p>That branch contained some other things that could interfere with
      test results. I started
      <a class="moz-txt-link-freetext" href="https://invent.kde.org/plasma/kwin/-/jobs/3959078">https://invent.kde.org/plasma/kwin/-/jobs/3959078</a> and yeah it
      looks like test run times are back to the Jan 29-30th level. </p>
    <p>So, it seems like fast_unwind_on_malloc=0 is the culprit then?</p>
    <p>Regards,<br>
      Vlad</p>
    <p><br>
    </p>
    <blockquote type="cite"
cite="mid:CA+XidOGtqOjws-+14LYwajipHpMBUCNSjSz8Y-2LELGi7yOuow@mail.gmail.com">
      <div dir="ltr">
        <div class="gmail_quote gmail_quote_container">
          <div>        Start  61: kwin-testOutputChanges<br>
             61/158 Test  #61: kwin-testOutputChanges
            .............................   Passed   19.58 sec</div>
          <div> </div>
          <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div dir="ltr">
              <div class="gmail_quote">
                <div> </div>
                <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                  <br>
                  Either way, the current state of CI is not great.
                  Hypothetically, test <br>
                  timeouts can be increased but QSignalSpy's have
                  hardcoded timeouts that <br>
                  can be too low for the current CI. And in case of
                  kwin, 5 minutes for a <br>
                  test is simply not a workable thing.<br>
                  <br>
                  Regards,<br>
                  Vlad<br>
                </blockquote>
                <div><br>
                </div>
                <div>Thanks,</div>
                <div>Ben </div>
              </div>
            </div>
          </blockquote>
          <div><br>
          </div>
          <div>Cheers,</div>
          <div>Ben </div>
        </div>
      </div>
    </blockquote>
  </body>
</html>