<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 - [X86] Failure to use movss/movsd/movq/movd when element 0 of a vector load is extracted and inserted into element 0 of a zero vector"
href="https://bugs.llvm.org/show_bug.cgi?id=39257">39257</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[X86] Failure to use movss/movsd/movq/movd when element 0 of a vector load is extracted and inserted into element 0 of a zero vector
</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>Backend: X86
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>craig.topper@gmail.com
</td>
</tr>
<tr>
<th>CC</th>
<td>llvm-bugs@lists.llvm.org
</td>
</tr></table>
<p>
<div>
<pre>See the 4 tests in vec_extract-avx.ll such as this
define void @legal_vzmovl_2f32_8f32(<2 x float>* %in, <8 x float>* %out) {
; X32-LABEL: legal_vzmovl_2f32_8f32:
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X32-NEXT: vxorps %xmm1, %xmm1, %xmm1
; X32-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; X32-NEXT: vmovaps %ymm0, (%eax)
; X32-NEXT: vzeroupper
; X32-NEXT: retl
;
; X64-LABEL: legal_vzmovl_2f32_8f32:
; X64: # %bb.0:
; X64-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
; X64-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; X64-NEXT: vmovaps %ymm0, (%rsi)
; X64-NEXT: vzeroupper
; X64-NEXT: retq
%ld = load <2 x float>, <2 x float>* %in, align 8
%ext = extractelement <2 x float> %ld, i64 0
%ins = insertelement <8 x float> <float undef, float 0.0, float 0.0, float
0.0, float 0.0, float 0.0, float 0.0, float 0.0>, float %ext, i64 0
store <8 x float> %ins, <8 x float>* %out, align 32
ret void
}
The load size really doesn't seem to matter. The same thing happens if you use
a 8xf32 load.</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>