From 1fd553c9345ff9f1f32b1256bae0f2d63ec3ae38 Mon Sep 17 00:00:00 2001
From: Guozhi Wei <carrot@google.com>
Date: Mon, 12 Dec 2016 22:09:02 +0000
Subject: [PPC] Prefer direct move on power8 if load 1 or 2 bytes to VSR

Power8 has MTVSRWZ but no LXSIBZX/LXSIHZX, so move 1 or 2 bytes to VSR through MTVSRWZ is much faster than store the extended value into stack and load it with LXSIWZX.
This patch fixes pr31144.

Differential Revision: https://reviews.llvm.org/D27287

llvm-svn: 289473
---
 llvm/test/CodeGen/PowerPC/pr31144.ll | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
 create mode 100644 llvm/test/CodeGen/PowerPC/pr31144.ll

(limited to 'llvm/test/CodeGen/PowerPC')

diff --git a/llvm/test/CodeGen/PowerPC/pr31144.ll b/llvm/test/CodeGen/PowerPC/pr31144.ll
new file mode 100644
index 00000000000..dc48d7cdfc3
--- /dev/null
+++ b/llvm/test/CodeGen/PowerPC/pr31144.ll
@@ -0,0 +1,26 @@
+; RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 -mattr=+vsx < %s | FileCheck %s --implicit-check-not lxsiwzx
+
+declare void @bar(double)
+
+define void @foo1(i8* %p) {
+entry:
+  %0 = load i8, i8* %p, align 1
+  %conv = uitofp i8 %0 to double
+  call void @bar(double %conv)
+  ret void
+
+; CHECK-LABEL: @foo1
+; CHECK:     mtvsrwz
+}
+
+define void @foo2(i16* %p) {
+entry:
+  %0 = load i16, i16* %p, align 2
+  %conv = uitofp i16 %0 to double
+  call void @bar(double %conv)
+  ret void
+
+; CHECK-LABEL: @foo2
+; CHECK:       mtvsrwz
+}
+
-- 
cgit v1.2.3