<html>
    <head>
      <base href="https://bugs.freedesktop.org/">
    </head>
    <body>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - vega56 undervolting/overclocking voltage issues"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=109887#c8">Comment # 8</a>
              on <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - vega56 undervolting/overclocking voltage issues"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=109887">bug 109887</a>
              from <span class="vcard"><a class="email" href="mailto:asheldon55@gmail.com" title="Andrew Sheldon <asheldon55@gmail.com>"> <span class="fn">Andrew Sheldon</span></a>
</span></b>
        <pre>I also can confirm the problem, and it seems to have gotten worse since
5.3.0-rcX.

In past kernels, you could kind of work around it by setting slightly less
conservative undervolts and it would work. If you go past a certain point (a
point that works fine if the pp table is overridden), it would wrap around to
1.2V. For reference this around 950mv at state 6/7.

Now with 5.3, even that same relatively conservative undervolt would
immediately jump to 1.2V under load.

As hagar-dunor suggested, manually overriding the entire PP table, works around
the problem.</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>