<html>
<head>
<base href="https://bugs.llvm.org/">
</head>
<body><span class="vcard"><a class="email" href="mailto:efriedma@quicinc.com" title="Eli Friedman <efriedma@quicinc.com>"> <span class="fn">Eli Friedman</span></a>
</span> changed
<a class="bz_bug_link
bz_status_REOPENED "
title="REOPENED - [AArch64] Missed CCMP opportunity"
href="https://bugs.llvm.org/show_bug.cgi?id=26084">bug 26084</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>RESOLVED
</td>
<td>REOPENED
</td>
</tr>
<tr>
<td style="text-align:right;">CC</td>
<td>
</td>
<td>efriedma@quicinc.com, florian_hahn@apple.com, sjoerd.meijer@arm.com
</td>
</tr>
<tr>
<td style="text-align:right;">Resolution</td>
<td>FIXED
</td>
<td>---
</td>
</tr></table>
<p>
<div>
<b><a class="bz_bug_link
bz_status_REOPENED "
title="REOPENED - [AArch64] Missed CCMP opportunity"
href="https://bugs.llvm.org/show_bug.cgi?id=26084#c4">Comment # 4</a>
on <a class="bz_bug_link
bz_status_REOPENED "
title="REOPENED - [AArch64] Missed CCMP opportunity"
href="https://bugs.llvm.org/show_bug.cgi?id=26084">bug 26084</a>
from <span class="vcard"><a class="email" href="mailto:efriedma@quicinc.com" title="Eli Friedman <efriedma@quicinc.com>"> <span class="fn">Eli Friedman</span></a>
</span></b>
<pre>The patch in question was reverted, and apparently nobody has looked into this
since.
I briefly tried rebasing the patch. It basically works, but the "custom"
markings have unintended consequences, and as noted in the revert, the
DAGCombine changes are suspect. (One of those consequences is suppressing
add->or conversion after legalization; filed <a class="bz_bug_link
bz_status_NEW "
title="NEW - AArch64 missed optimization involving "or(frameindex, n)""
href="show_bug.cgi?id=46151">bug 46151</a> and <a class="bz_bug_link
bz_status_NEW "
title="NEW - aarch64 "bfm" matching fails with known-zero bits"
href="show_bug.cgi?id=46152">bug 46152</a>.)
I don't think marking integer OR for custom lowering is an appropriate
solution; there are too many potential weird consequences in terms of
suppressing target-independent DAGCombines. Probably should be a DAGCombine
itself. (Not sure if it would run before or after legalization. Before,
checking profitability is hard. After, we have to write some new code to dig
into target-specific compare nodes.)</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>