<html>
<head>
<base href="https://bugs.freedesktop.org/">
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - [i915] Feature request: Add support for fencing for PRIME setups"
href="https://bugs.freedesktop.org/show_bug.cgi?id=95472#c13">Comment # 13</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - [i915] Feature request: Add support for fencing for PRIME setups"
href="https://bugs.freedesktop.org/show_bug.cgi?id=95472">bug 95472</a>
from <span class="vcard"><a class="email" href="mailto:chris@chris-wilson.co.uk" title="Chris Wilson <chris@chris-wilson.co.uk>"> <span class="fn">Chris Wilson</span></a>
</span></b>
<pre>Thinking through the DRI3 chain.
GL/dgpu client renders into back buffer
-> SwapBuffers, GL/dgpu copies into linear buffer (dma-buf pixmap) and calls
PresentPixmap
-> X/igfx either flips the fullscreen dma-buf pixmap, or copies from the
dma-buf pixmap into the frontbuffer
Here we should have a write/exclusive fence installed by the client and that
should be waited upon by i915.ko inside X. The copy/flip are different call
paths inside i915.ko, but both should be waiting for read access on the fence.
Does the behaviour differ if the rendered app is fullscreen (and make sure you
have fullscreen unredirection enabled)?</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are on the CC list for the bug.</li>
<li>You are the QA Contact for the bug.</li>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>