<html>
<head>
<base href="https://bugs.llvm.org/">
</head>
<body><span class="vcard"><a class="email" href="mailto:llvm-dev@redking.me.uk" title="Simon Pilgrim <llvm-dev@redking.me.uk>"> <span class="fn">Simon Pilgrim</span></a>
</span> changed
<a class="bz_bug_link
bz_status_RESOLVED bz_closed"
title="RESOLVED FIXED - No vectorization - should use pmaddwd"
href="https://bugs.llvm.org/show_bug.cgi?id=47437">bug 47437</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;">Status</td>
<td>NEW
</td>
<td>RESOLVED
</td>
</tr>
<tr>
<td style="text-align:right;">Fixed By Commit(s)</td>
<td>
</td>
<td>6ec644e2157d63acccacd877cf08c68d5836c169
</td>
</tr>
<tr>
<td style="text-align:right;">Resolution</td>
<td>---
</td>
<td>FIXED
</td>
</tr></table>
<p>
<div>
<b><a class="bz_bug_link
bz_status_RESOLVED bz_closed"
title="RESOLVED FIXED - No vectorization - should use pmaddwd"
href="https://bugs.llvm.org/show_bug.cgi?id=47437#c4">Comment # 4</a>
on <a class="bz_bug_link
bz_status_RESOLVED bz_closed"
title="RESOLVED FIXED - No vectorization - should use pmaddwd"
href="https://bugs.llvm.org/show_bug.cgi?id=47437">bug 47437</a>
from <span class="vcard"><a class="email" href="mailto:llvm-dev@redking.me.uk" title="Simon Pilgrim <llvm-dev@redking.me.uk>"> <span class="fn">Simon Pilgrim</span></a>
</span></b>
<pre>(In reply to Simon Pilgrim from <a href="show_bug.cgi?id=47437#c3">comment #3</a>)
<span class="quote">> (In reply to David Bolvansky from <a href="show_bug.cgi?id=47437#c2">comment #2</a>)
> > Thanks for improvements, with AVX LLVM 14 produces vpmaddwd, not yet with
> > plain "-O3".
>
> It should in fact work with most SSE41+ targets (although not SLM which has
> some custom costs that push it just over the threshold).
>
> The next step is to improve pre-AVX2 interleaved costs which should allow
> the LV costs analysis to see this as beneficial.
>
> I've started the process at <a href="https://reviews.llvm.org/D111822">https://reviews.llvm.org/D111822</a></span >
<a href="https://reviews.llvm.org/D111938">https://reviews.llvm.org/D111938</a> should fix this whenever interleaving is
enabled.
-march=atom is the only case where we don't enable interleaving on x86,
although we probably should now that the atom model and costs have been
improved.</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are on the CC list for the bug.</li>
</ul>
</body>
</html>