mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-01 16:33:37 +01:00
1025d1eb3b
the patterns already there to be more strict regarding the predicate. This fixes PR10558 llvm-svn: 137100
43 lines
1.4 KiB
LLVM
43 lines
1.4 KiB
LLVM
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=corei7-avx -mattr=+avx | FileCheck %s
|
|
|
|
; CHECK: vmovaps
|
|
; CHECK: vmovaps
|
|
; CHECK: vmovapd
|
|
; CHECK: vmovapd
|
|
; CHECK: vmovaps
|
|
; CHECK: vmovaps
|
|
define void @test_256_load(double* nocapture %d, float* nocapture %f, <4 x i64>* nocapture %i) nounwind uwtable ssp {
|
|
entry:
|
|
%0 = bitcast double* %d to <4 x double>*
|
|
%tmp1.i = load <4 x double>* %0, align 32
|
|
%1 = bitcast float* %f to <8 x float>*
|
|
%tmp1.i17 = load <8 x float>* %1, align 32
|
|
%tmp1.i16 = load <4 x i64>* %i, align 32
|
|
tail call void @dummy(<4 x double> %tmp1.i, <8 x float> %tmp1.i17, <4 x i64> %tmp1.i16) nounwind
|
|
store <4 x double> %tmp1.i, <4 x double>* %0, align 32
|
|
store <8 x float> %tmp1.i17, <8 x float>* %1, align 32
|
|
store <4 x i64> %tmp1.i16, <4 x i64>* %i, align 32
|
|
ret void
|
|
}
|
|
|
|
declare void @dummy(<4 x double>, <8 x float>, <4 x i64>)
|
|
|
|
;;
|
|
;; The two tests below check that we must fold load + scalar_to_vector
|
|
;; + ins_subvec+ zext into only a single vmovss or vmovsd
|
|
|
|
; CHECK: vmovss (%
|
|
define <8 x float> @mov00(<8 x float> %v, float * %ptr) nounwind {
|
|
%val = load float* %ptr
|
|
%i0 = insertelement <8 x float> zeroinitializer, float %val, i32 0
|
|
ret <8 x float> %i0
|
|
}
|
|
|
|
; CHECK: vmovsd (%
|
|
define <4 x double> @mov01(<4 x double> %v, double * %ptr) nounwind {
|
|
%val = load double* %ptr
|
|
%i0 = insertelement <4 x double> zeroinitializer, double %val, i32 0
|
|
ret <4 x double> %i0
|
|
}
|
|
|