<html>
<head>
<base href="https://bugs.freedesktop.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - unsupported call to function erf when running GROMACS with OpenCL"
href="https://bugs.freedesktop.org/show_bug.cgi?id=93375#c2">Comment # 2</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - unsupported call to function erf when running GROMACS with OpenCL"
href="https://bugs.freedesktop.org/show_bug.cgi?id=93375">bug 93375</a>
from <span class="vcard"><a class="email" href="mailto:rivanvx@gmail.com" title="Vedran Miletić <rivanvx@gmail.com>"> <span class="fn">Vedran Miletić</span></a>
</span></b>
<pre>(In reply to Aaron Watry from <a href="show_bug.cgi?id=93375#c1">comment #1</a>)
<span class="quote">> Looks like libclc does not yet implement the erf function.</span >
Neat, the approach you suggest seems to be simpler than the one I had in mind,
analogous to ldexp <a href="http://reviews.llvm.org/D14990">http://reviews.llvm.org/D14990</a>
<span class="quote">> Note that it looks like the AMD implementation of erf requires doubles if
> subnormals are NOT supported... If that's an issue for someone, we'll need a
> workaround for that.</span >
erff() from GROMACS works without doubles:
<a href="https://github.com/gromacs/gromacs/blob/fab0108c88e441d891d838bb81ca879ff9708317/src/gromacs/math/utilities.cpp">https://github.com/gromacs/gromacs/blob/fab0108c88e441d891d838bb81ca879ff9708317/src/gromacs/math/utilities.cpp</a>
Not sure if it's usable in this form, I would have to look.</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>