<html>
<head>
<base href="https://bugs.llvm.org/">
</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 - xxhash64 probably not the best choice for SHF_MERGE merging"
href="https://bugs.llvm.org/show_bug.cgi?id=37029">37029</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>xxhash64 probably not the best choice for SHF_MERGE merging
</td>
</tr>
<tr>
<th>Product</th>
<td>lld
</td>
</tr>
<tr>
<th>Version</th>
<td>unspecified
</td>
</tr>
<tr>
<th>Hardware</th>
<td>PC
</td>
</tr>
<tr>
<th>OS</th>
<td>Linux
</td>
</tr>
<tr>
<th>Status</th>
<td>NEW
</td>
</tr>
<tr>
<th>Severity</th>
<td>enhancement
</td>
</tr>
<tr>
<th>Priority</th>
<td>P
</td>
</tr>
<tr>
<th>Component</th>
<td>ELF
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>rafael@espindo.la
</td>
</tr>
<tr>
<th>CC</th>
<td>llvm-bugs@lists.llvm.org
</td>
</tr></table>
<p>
<div>
<pre>As a quick experiment I replaced the two calls to xxhash64 in splitIntoPieces
with hash_value. This resulted in a 4% speedup when linking scylla.
While xxHash64 was 13.3% of the runtime of master, hash_combine is just 6.6% of
the patch.
We can't just use hash_combine since it produces different results on 32 bit
hosts, but refactoring the algorithm to have the option of a stable result is
probably worth it.</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>