<html>
<head>
<base href="https://bugs.freedesktop.org/">
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - Assertion failed: size_log2 <= ANV_MAX_STATE_SIZE_LOG2 While compiling large compute shader"
href="https://bugs.freedesktop.org/show_bug.cgi?id=100092#c4">Comment # 4</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - Assertion failed: size_log2 <= ANV_MAX_STATE_SIZE_LOG2 While compiling large compute shader"
href="https://bugs.freedesktop.org/show_bug.cgi?id=100092">bug 100092</a>
from <span class="vcard"><a class="email" href="mailto:registration-bugzilla@schwa.ca" title="Josh Gargus <registration-bugzilla@schwa.ca>"> <span class="fn">Josh Gargus</span></a>
</span></b>
<pre>Thanks Jason!
In general on NVIDIA, it seems faster to store values in registers than shared
memory, *unless* there is not enough register space and it needs to spill the
data elsewhere (to main GPU memory? to some other cache? I have no idea).
In this case, it seems that making roi shared reduces the register pressure
enough that depth can fit in registers; as a result, making depth shared
reduces performance.
At least, that is my interpretation.</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
<li>You are the QA Contact for the bug.</li>
</ul>
</body>
</html>