<html>
<head>
<base href="https://bugs.freedesktop.org/">
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - Regression in Mesa 17 on s390x (zSystems)"
href="https://bugs.freedesktop.org/show_bug.cgi?id=100613#c26">Comment # 26</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - Regression in Mesa 17 on s390x (zSystems)"
href="https://bugs.freedesktop.org/show_bug.cgi?id=100613">bug 100613</a>
from <span class="vcard"><a class="email" href="mailto:robclark@freedesktop.org" title="Rob Clark <robclark@freedesktop.org>"> <span class="fn">Rob Clark</span></a>
</span></b>
<pre>(In reply to Ben Crocker from <a href="show_bug.cgi?id=100613#c25">comment #25</a>)
>
<span class="quote">> Regarding Ray's specific comment about getting scalar fetch to work
> with "sufficient twiddling," I think it's perfectly acceptable to
> introduce extra operations, as long as we restrict the extra
> operations to the big-endian path. PPC64 (LE or BE) is fast enough so
> that any performance impact will be negligible; S390 is less fast, but
> I imagine production machines with more memory than the one we
> experimented on here are fast enough.</span >
drive-by comment.. unless llvm is just rubbish at optimization, I don't think
saving a few operations in the front-end IR building should be that important,
even for LE. But we have shader-db so it should be possible to prove/disprove
that theory. (Not sure if llvmpipe is instrumented for shader-db but if not
that should be easy to solve.)</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 the assignee for the bug.</li>
</ul>
</body>
</html>