<html>
    <head>
      <base href="https://bugs.freedesktop.org/">
    </head>
    <body>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - AMD Navi10 GPU powerplay issues when using two DisplayPort connectors"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=111913#c12">Comment # 12</a>
              on <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - AMD Navi10 GPU powerplay issues when using two DisplayPort connectors"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=111913">bug 111913</a>
              from <span class="vcard"><a class="email" href="mailto:stefan@rehm.email" title="Stefan Rehm <stefan@rehm.email>"> <span class="fn">Stefan Rehm</span></a>
</span></b>
        <pre>If I change

dcn2_0_soc.dram_clock_change_latency_us

in "rivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c" from 404.0 to 10.0
(the value used in kernel 5.3) the messages disappear and the system behaves
normal again. However, as long as

/sys/class/drm/card0/device/power_dpm_force_performance_level

is set to "auto", I am now seeing massive flickering. Forcing it to low or high
fixes that.

According to the sources for kernel 5.3 the value of 10.0 for
dram_clock_change_latency is a hack. Can anyone elaborate on this?</pre>
        </div>
      </p>


      <hr>
      <span>You are receiving this mail because:</span>

      <ul>
          <li>You are the assignee for the bug.</li>
      </ul>
    </body>
</html>