2012-04-27 07:11:58 +00:00
|
|
|
; RUN: llc < %s -mtriple=i386-apple-darwin -mattr=+sse2 -mcpu=nehalem | FileCheck %s
|
Optimize splat of a scalar load into a shuffle of a vector load when it's legal. e.g.
vector_shuffle (scalar_to_vector (i32 load (ptr + 4))), undef, <0, 0, 0, 0>
=>
vector_shuffle (v4i32 load ptr), undef, <1, 1, 1, 1>
iff ptr is 16-byte aligned (or can be made into 16-byte aligned).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90984 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-09 21:00:30 +00:00
|
|
|
; rdar://7434544
|
|
|
|
|
2011-11-23 07:13:56 +00:00
|
|
|
define <2 x i64> @t2() nounwind {
|
Optimize splat of a scalar load into a shuffle of a vector load when it's legal. e.g.
vector_shuffle (scalar_to_vector (i32 load (ptr + 4))), undef, <0, 0, 0, 0>
=>
vector_shuffle (v4i32 load ptr), undef, <1, 1, 1, 1>
iff ptr is 16-byte aligned (or can be made into 16-byte aligned).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90984 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-09 21:00:30 +00:00
|
|
|
entry:
|
2013-07-14 06:24:09 +00:00
|
|
|
; CHECK-LABEL: t2:
|
Optimize splat of a scalar load into a shuffle of a vector load when it's legal. e.g.
vector_shuffle (scalar_to_vector (i32 load (ptr + 4))), undef, <0, 0, 0, 0>
=>
vector_shuffle (v4i32 load ptr), undef, <1, 1, 1, 1>
iff ptr is 16-byte aligned (or can be made into 16-byte aligned).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90984 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-09 21:00:30 +00:00
|
|
|
; CHECK: pshufd $85, (%esp), %xmm0
|
|
|
|
%array = alloca [8 x float], align 4
|
|
|
|
%arrayidx = getelementptr inbounds [8 x float]* %array, i32 0, i32 1
|
|
|
|
%tmp2 = load float* %arrayidx
|
|
|
|
%vecinit = insertelement <4 x float> undef, float %tmp2, i32 0
|
|
|
|
%vecinit5 = insertelement <4 x float> %vecinit, float %tmp2, i32 1
|
|
|
|
%vecinit7 = insertelement <4 x float> %vecinit5, float %tmp2, i32 2
|
|
|
|
%vecinit9 = insertelement <4 x float> %vecinit7, float %tmp2, i32 3
|
|
|
|
%0 = bitcast <4 x float> %vecinit9 to <2 x i64>
|
|
|
|
ret <2 x i64> %0
|
|
|
|
}
|