<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][SSE] Masked memory op expansion could use MOVMSK to simplify mask handling"
href="https://bugs.llvm.org/show_bug.cgi?id=39665">39665</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[X86][SSE] Masked memory op expansion could use MOVMSK to simplify mask handling
</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>llvm-dev@redking.me.uk
</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>define <2 x double> @load_v2f64_v2i64(<2 x i64> %trigger, <2 x double>* %addr,
<2 x double> %dst) {
; SSE42-LABEL: load_v2f64_v2i64:
; SSE42: ## %bb.0:
; SSE42-NEXT: pxor %xmm2, %xmm2
; SSE42-NEXT: pcmpeqq %xmm0, %xmm2
; SSE42-NEXT: pextrb $0, %xmm2, %eax
; SSE42-NEXT: testb $1, %al
; SSE42-NEXT: je LBB1_2
; SSE42-NEXT: ## %bb.1: ## %cond.load
; SSE42-NEXT: movlpd {{.*#+}} xmm1 = mem[0],xmm1[1]
; SSE42-NEXT: LBB1_2: ## %else
; SSE42-NEXT: pextrb $8, %xmm2, %eax
; SSE42-NEXT: testb $1, %al
; SSE42-NEXT: je LBB1_4
; SSE42-NEXT: ## %bb.3: ## %cond.load1
; SSE42-NEXT: movhpd {{.*#+}} xmm1 = xmm1[0],mem[0]
; SSE42-NEXT: LBB1_4: ## %else2
; SSE42-NEXT: movapd %xmm1, %xmm0
; SSE42-NEXT: retq
%mask = icmp eq <2 x i64> %trigger, zeroinitializer
%res = call <2 x double> @llvm.masked.load.v2f64.p0v2f64(<2 x double>* %addr,
i32 4, <2 x i1>%mask, <2 x double>%dst)
ret <2 x double> %res
}
declare <2 x double> @llvm.masked.load.v2f64.p0v2f64(<2 x double>*, i32, <2 x
i1>, <2 x double>)
This can replace all the pextrb with a single movmsk to something like:
pxor %xmm2, %xmm2
pcmpeqq %xmm0, %xmm2
movmskpd %xmm2, %eax
testb $1, %al
jne LBB1_2
## %bb.1: ## %cond.load
movlpd {{.*#+}} xmm1 = mem[0],xmm1[1]
LBB1_2: ## %else
testb $2, %al
jne LBB1_4
## %bb.3: ## %cond.load1
movhpd {{.*#+}} xmm1 = xmm1[0],mem[0]
LBB1_4: ## %else2
movapd %xmm1, %xmm0
retq</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>