<html>
<head>
<base href="https://bugs.freedesktop.org/">
</head>
<body><span class="vcard"><a class="email" href="mailto:daniele.ceraolospurio@intel.com" title="Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>"> <span class="fn">Daniele Ceraolo Spurio</span></a>
</span> changed
<a class="bz_bug_link
bz_status_NEW "
title="NEW - [CI][BAT][ICL-GUC] igt@i915_selftest@live_hangcheck - dmesg-warn - GuC status: 0x8002f076, MIA core expected to be in reset"
href="https://bugs.freedesktop.org/show_bug.cgi?id=111674">bug 111674</a>
<br>
<table border="1" cellspacing="0" cellpadding="8">
<tr>
<th>What</th>
<th>Removed</th>
<th>Added</th>
</tr>
<tr>
<td style="text-align:right;">Severity</td>
<td>not set
</td>
<td>major
</td>
</tr>
<tr>
<td style="text-align:right;">Priority</td>
<td>not set
</td>
<td>low
</td>
</tr></table>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - [CI][BAT][ICL-GUC] igt@i915_selftest@live_hangcheck - dmesg-warn - GuC status: 0x8002f076, MIA core expected to be in reset"
href="https://bugs.freedesktop.org/show_bug.cgi?id=111674#c3">Comment # 3</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - [CI][BAT][ICL-GUC] igt@i915_selftest@live_hangcheck - dmesg-warn - GuC status: 0x8002f076, MIA core expected to be in reset"
href="https://bugs.freedesktop.org/show_bug.cgi?id=111674">bug 111674</a>
from <span class="vcard"><a class="email" href="mailto:daniele.ceraolospurio@intel.com" title="Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>"> <span class="fn">Daniele Ceraolo Spurio</span></a>
</span></b>
<pre>It looks like something went wrong in the HW.
The test does lots of full gt resets; i915 resets the GuC before doing a full
gt reset and in one case the GuC reset seems to have gone wrong. The HW
reported a successful reset (confirmed by the lack of error logs for this) but
the GUC_STATUS register, which we check right after the reset, has not been
correctly reset and contains a status that makes no sense: one part indicates
that GuC is executing the bootrom, while another part that GuC loading has
completed and that GuC is in halt state, both of which are not something that
can happen immediately after the reset has completed.
We only reset and reload GuC after a resume or a GT reset and in both of those
cases we reset the GuC at least twice (once in GT reset/sanitize and once
immediately before load), so we can handle one of the resets not correctly
resetting all the regs. If all the resets fail then we could end up failing to
load GuC, but this is very unlikely.</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the QA Contact for the bug.</li>
<li>You are on the CC list for the bug.</li>
</ul>
</body>
</html>