<html>
<head>
<base href="http://llvm.org/bugs/" />
</head>
<body><span class="vcard"><a class="email" href="mailto:atrick@apple.com" title="Andrew Trick <atrick@apple.com>"> <span class="fn">Andrew Trick</span></a>
</span> changed
<a class="bz_bug_link
bz_status_RESOLVED bz_closed"
title="RESOLVED FIXED - clang produces incorrect code with loop/expression at -O2"
href="http://llvm.org/bugs/show_bug.cgi?id=16130">bug 16130</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>REOPENED
</td>
<td>RESOLVED
</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 - clang produces incorrect code with loop/expression at -O2"
href="http://llvm.org/bugs/show_bug.cgi?id=16130#c16">Comment # 16</a>
on <a class="bz_bug_link
bz_status_RESOLVED bz_closed"
title="RESOLVED FIXED - clang produces incorrect code with loop/expression at -O2"
href="http://llvm.org/bugs/show_bug.cgi?id=16130">bug 16130</a>
from <span class="vcard"><a class="email" href="mailto:atrick@apple.com" title="Andrew Trick <atrick@apple.com>"> <span class="fn">Andrew Trick</span></a>
</span></b>
<pre>The second half of the fix is r183060.
This handles the case when the original source loop is not inlined and we have
multiple loop exits. We normally handle multiple exits conservatively, but loop
unrolling was making an assumption that the loop latch test provided an upper
bound on trip count. I changed it to be conservative.
We could definitely do better at computing trip counts. Tracking that
internally...
rdar:14038809: Optimize trip count computation for multi-exit loops.</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>