<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 - [DAG] DAGTypeLegalizer::GenWidenVectorTruncStores - possible copy+paste typo"
href="https://bugs.llvm.org/show_bug.cgi?id=42046">42046</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[DAG] DAGTypeLegalizer::GenWidenVectorTruncStores - possible copy+paste typo
</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>enhancement
</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>llvm-dev@redking.me.uk
</td>
</tr>
<tr>
<th>CC</th>
<td>craig.topper@gmail.com, llvm-bugs@lists.llvm.org, llvm-dev@ndave.org, spatel+llvm@rotateright.com
</td>
</tr></table>
<p>
<div>
<pre>unsigned NumElts = StVT.getVectorNumElements();
SDValue EOp = DAG.getNode(
ISD::EXTRACT_VECTOR_ELT, dl, ValEltVT, ValOp,
DAG.getConstant(0, dl, TLI.getVectorIdxTy(DAG.getDataLayout())));
StChain.push_back(DAG.getTruncStore(Chain, dl, EOp, BasePtr,
ST->getPointerInfo(), StEltVT, Align,
MMOFlags, AAInfo));
unsigned Offset = Increment;
for (unsigned i=1; i < NumElts; ++i, Offset += Increment) {
SDValue NewBasePtr = DAG.getObjectPtrOffset(dl, BasePtr, Offset);
SDValue EOp = DAG.getNode(
ISD::EXTRACT_VECTOR_ELT, dl, ValEltVT, ValOp,
DAG.getConstant(0, dl, TLI.getVectorIdxTy(DAG.getDataLayout())));
StChain.push_back(DAG.getTruncStore(
Chain, dl, EOp, NewBasePtr, ST->getPointerInfo().getWithOffset(Offset),
StEltVT, MinAlign(Align, Offset), MMOFlags, AAInfo));
}
The ISD::EXTRACT_VECTOR_ELT in the loop is repeatedly extracting the 0'th
vector element - shouldn't it be be the i'th element index?
According to <a href="http://lab.llvm.org:8080/coverage/coverage-reports/llvm/coverage">http://lab.llvm.org:8080/coverage/coverage-reports/llvm/coverage</a>
this code isn't tested at all.
I only noticed this because 'SDValue EOp' was causing a Wshadow warning in
cppcheck.</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>