<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 - Missing template instantiations from HIP device code"
href="https://bugs.llvm.org/show_bug.cgi?id=44416">44416</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>Missing template instantiations from HIP device code
</td>
</tr>
<tr>
<th>Product</th>
<td>clang
</td>
</tr>
<tr>
<th>Version</th>
<td>trunk
</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>C++
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedclangbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>jonathanchesterfield@gmail.com
</td>
</tr>
<tr>
<th>CC</th>
<td>blitzrakete@gmail.com, dgregor@apple.com, erik.pilkington@gmail.com, llvm-bugs@lists.llvm.org, richard-llvm@metafoo.co.uk
</td>
</tr></table>
<p>
<div>
<pre>A template instantiation in hip code compiled with cuda-device-only is missing
from the generated bitcode.
Component is HIP, but there is no HIP on the drop down list so marking as C++.
<a href="https://godbolt.org/z/GB9MY9">https://godbolt.org/z/GB9MY9</a>
clang atomic.hip --cuda-device-only -nogpulib --cuda-gpu-arch=gfx906 -emit-llvm
#define DEV __attribute__((device))
// Function emitted as expected
DEV int atomic_inc_int(int *x) {
return __atomic_fetch_add(x, 1, __ATOMIC_SEQ_CST);
}
// Template definition
template <typename T> DEV T atomic_inc(T *x) {
return __atomic_fetch_add(x, 1, __ATOMIC_SEQ_CST);
}
// Explict instantiation, not emitted
template DEV int atomic_inc<int>(int *);
// Use of implicit instantiation, not emitted
auto *leak_addr = &atomic_inc<int>;
Results in IR
target datalayout =
"e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5-ni:7"
target triple = "amdgcn-amd-amdhsa"
define hidden i32 @_Z14atomic_inc_intPi(i32* nocapture %0) local_unnamed_addr
#0 {
%2 = atomicrmw add i32* %0, i32 1 seq_cst
ret i32 %2
}
attributes #0 = { ... }
!llvm.module.flags = !{!0, !1}
!llvm.ident = !{!2}
!0 = !{i32 1, !"wchar_size", i32 4}
!1 = !{i32 7, !"PIC Level", i32 1}
!2 = !{!"clang version 10.0.0 (<a href="https://github.com/llvm/llvm-project.git">https://github.com/llvm/llvm-project.git</a>
6185dc0eb3ad35e1f85f2ab1038ca978563099f4)"}
Also raised on a downstream tracker before realising it reproduces on trunk
(<a href="https://github.com/ROCm-Developer-Tools/aomp/issues/67">https://github.com/ROCm-Developer-Tools/aomp/issues/67</a>)</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>