<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 - vpermq+vpshufb intrinsics pessimized into 2vpshufb+vpermq+blend"
href="https://bugs.llvm.org/show_bug.cgi?id=47262">47262</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>vpermq+vpshufb intrinsics pessimized into 2vpshufb+vpermq+blend
</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>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>Backend: X86
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>tellowkrinkle@gmail.com
</td>
</tr>
<tr>
<th>CC</th>
<td>craig.topper@gmail.com, llvm-bugs@lists.llvm.org, llvm-dev@redking.me.uk, spatel+llvm@rotateright.com
</td>
</tr></table>
<p>
<div>
<pre>Godbolt link: <a href="https://gcc.godbolt.org/z/PdheW9">https://gcc.godbolt.org/z/PdheW9</a>
The following code, when compiled with `clang -march=haswell`:
#include <immintrin.h>
__m256 test(__m256 in) {
__m128 mask = _mm_setr_epi8(0, 4, 8, 12, 1, 5, 9, 13, 2, 6, 10, 14, 3, 7,
11, 15);
__m256 tmp = _mm256_permute4x64_epi64(in, _MM_SHUFFLE(3, 1, 2, 0));
return _mm256_shuffle_epi8(tmp, _mm256_broadcastsi128_si256(mask));
}
unexpectedly produces the following assembly:
vpshufb ymm1, ymm0, ymmword ptr [rip + .shuffleMask1]
vpermq ymm0, ymm0, 78 # ymm0 = ymm0[2,3,0,1]
vpshufb ymm0, ymm0, ymmword ptr [rip + .shuffleMask2]
vmovdqa ymm2, ymmword ptr [rip + .blendMask]
vpblendvb ymm0, ymm0, ymm1, ymm2
compared to the expected:
vpermq ymm0, ymm0, 216 # ymm0 = ymm0[0,2,1,3]
vpshufb ymm0, ymm0, ymmword ptr [rip + .shuffleMask]
The cause appears to be instruction selection merging the two shuffles, and
then failing to figure out a reasonable lowering, falling back to the final
generic "shuffle the two lanes separately and blend them together"
In general, I think any cross-lane byte shuffle where the number of elements
crossing lanes is divisible by 4 would be better lowered as a vpshufb to
gather, vperm[d|q] to cross lanes, and a final vpshufb to scatter</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>