<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 - [NVPTX] Wrong results or hang on GPU when using lastprivate() on SPMD construct with full runtime"
href="https://bugs.llvm.org/show_bug.cgi?id=39172">39172</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[NVPTX] Wrong results or hang on GPU when using lastprivate() on SPMD construct with full runtime
</td>
</tr>
<tr>
<th>Product</th>
<td>OpenMP
</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>Clang Compiler Support
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedclangbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>hahnjo@hahnjo.de
</td>
</tr>
<tr>
<th>CC</th>
<td>llvm-bugs@lists.llvm.org
</td>
</tr></table>
<p>
<div>
<pre>The following program uses an SPMD construct with lastprivate() clause and the
runtime has to be initialized because of schedule(dynamic):
#include <stdio.h>
#include <stdlib.h>
int main(int argc, char *argv[]) {
int last;
#pragma omp target teams distribute parallel for map(from: last)
lastprivate(last) schedule(dynamic)
for (int i = 0; i < 10000; i++) {
last = i;
}
printf("last = %d\n", last);
return EXIT_SUCCESS;
}
Compiled with current Clang trunk the application delivers wrong results (last
= 0) with -O0 and -O1 and hangs with -O2 and -O3. I think this is due to the
same problem: The generated code calls __kmpc_data_sharing_push_stack() to get
memory for storing "int last;" to implement lastprivate(). This works if the
runtime is uninitialized because there is a special case in libomptarget-nvptx
which returns the *same memory location for all threads* (see
<a href="https://reviews.llvm.org/D51875">https://reviews.llvm.org/D51875</a>).
However the original contract of __kmpc_data_sharing_push_stack() was to return
*unique memory for each thread* which explains the wrong result. For higher
optimization levels I'd guess that LLVM exploits undefined behaviour somewhere
which makes the application hang?
The only solution I can think of is to introduce new entry points in
libomptarget-nvptx with the required contract: Return the same memory location
for all calling threads in the same thread block. Opinions?</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>