<html>
<head>
<base href="https://llvm.org/bugs/" />
</head>
<body><table border="1" cellspacing="0" cellpadding="8">
<tr>
<th>Bug ID</th>
<td><a class="bz_bug_link
bz_status_NEW "
title="NEW --- - BFNZ not generated for simple loops"
href="https://llvm.org/bugs/show_bug.cgi?id=28363">28363</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>BFNZ not generated for simple loops
</td>
</tr>
<tr>
<th>Product</th>
<td>libraries
</td>
</tr>
<tr>
<th>Version</th>
<td>trunk
</td>
</tr>
<tr>
<th>Hardware</th>
<td>PC
</td>
</tr>
<tr>
<th>OS</th>
<td>Windows NT
</td>
</tr>
<tr>
<th>Status</th>
<td>NEW
</td>
</tr>
<tr>
<th>Severity</th>
<td>normal
</td>
</tr>
<tr>
<th>Priority</th>
<td>P
</td>
</tr>
<tr>
<th>Component</th>
<td>Backend: PowerPC
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>amehsan@ca.ibm.com
</td>
</tr>
<tr>
<th>CC</th>
<td>llvm-bugs@lists.llvm.org
</td>
</tr>
<tr>
<th>Classification</th>
<td>Unclassified
</td>
</tr></table>
<p>
<div>
<pre>For many loops, sometimes very simple ones like:
for (i = 0; i < m; i++)
sum += b[i]*a[i];
Where sum, a and b are single precision float (arrays), we do not generate
bdnz, and instead use, three instructions in the loop. (add, compare,
branch-conditional).
I looked at one of the loops like this and Scalar Evolution was not able to
compute loop iteration count. When the loop is not unrolled (compile with
-fno-unroll-loops) this is not an issue. Also for the residue loops of an
unrolled loop we generate bdnz.</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>