<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 --- - [Scheduler] Recent improvements lead to compile time bloat"
href="https://llvm.org/bugs/show_bug.cgi?id=26940">26940</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[Scheduler] Recent improvements lead to compile time bloat
</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>All
</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>Common Code Generator Code
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>bruno.cardoso@gmail.com
</td>
</tr>
<tr>
<th>CC</th>
<td>llvm-bugs@lists.llvm.org, paulsson@linux.vnet.ibm.com
</td>
</tr>
<tr>
<th>Classification</th>
<td>Unclassified
</td>
</tr></table>
<p>
<div>
<pre>Commit r259673 (“[ScheduleDAGInstrs::buildSchedGraph()] Handling of memory
dependecies rewritten”) introduced a huge slowdown in an internal benchmark:
=== Before r259673
$ clang -arch arm64 -Os -c t.mi
real 1389.12s
user 1314.71s => 22min
sys 14.92s
$ clang -arch x86_64 -Os -c t.mi
real 76.90s
user 75.46s => 1min15s
sys 0.92s
=== After r259673
real 10045.62s
user 9131.03s => 3h20min
sys 95.60s
real 2217.71s
user 2174.12s => 36min23s
sys 33.07s
ARM64 is a lot worse than x86, but compile time for both are pretty bad! Note
that before r259673 arm64 time was already bad (22min) but got even worse
after. Gonna try to reduce the testcase / provide a testcase with the same
reproducible slowdown factor.</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>