mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 19:52:54 +01:00
47b83fb85b
This improves the fix committed at revision 199683 adding the
following new target specific combine rules:
1) fold (v4i32: vselect <0,0,-1,-1>, A, B) ->
(v4i32 (bitcast (movsd (v2i64 (bitcast A)), (v2i64 (bitcast B))) ))
2) fold (v4f32: vselect <0,0,-1,-1>, A, B) ->
(v4f32 (bitcast (movsd (v2f64 (bitcast A)), (v2f64 (bitcast B))) ))
3) fold (v4i32: vselect <-1,-1,0,0>, A, B) ->
(v4i32 (bitcast (movsd (v2i64 (bitcast B)), (v2i64 (bitcast A))) ))
4) fold (v4f32: vselect <-1,-1,0,0>, A, B) ->
(v4f32 (bitcast (movsd (v2i64 (bitcast B)), (v2i64 (bitcast A))) ))
llvm-svn: 200324
34 lines
1020 B
LLVM
34 lines
1020 B
LLVM
; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -mattr=sse2 | FileCheck %s
|
|
|
|
define <4 x i32> @test1(<4 x i32> %A, <4 x i32> %B) {
|
|
%select = select <4 x i1><i1 true, i1 true, i1 false, i1 false>, <4 x i32> %A, <4 x i32> %B
|
|
ret <4 x i32> %select
|
|
}
|
|
; CHECK-LABEL: test1
|
|
; CHECK: movsd
|
|
; CHECK: ret
|
|
|
|
define <4 x i32> @test2(<4 x i32> %A, <4 x i32> %B) {
|
|
%select = select <4 x i1><i1 false, i1 false, i1 true, i1 true>, <4 x i32> %A, <4 x i32> %B
|
|
ret <4 x i32> %select
|
|
}
|
|
; CHECK-LABEL: test2
|
|
; CHECK: movsd
|
|
; CHECK-NEXT: ret
|
|
|
|
define <4 x float> @test3(<4 x float> %A, <4 x float> %B) {
|
|
%select = select <4 x i1><i1 true, i1 true, i1 false, i1 false>, <4 x float> %A, <4 x float> %B
|
|
ret <4 x float> %select
|
|
}
|
|
; CHECK-LABEL: test3
|
|
; CHECK: movsd
|
|
; CHECK: ret
|
|
|
|
define <4 x float> @test4(<4 x float> %A, <4 x float> %B) {
|
|
%select = select <4 x i1><i1 false, i1 false, i1 true, i1 true>, <4 x float> %A, <4 x float> %B
|
|
ret <4 x float> %select
|
|
}
|
|
; CHECK-LABEL: test4
|
|
; CHECK: movsd
|
|
; CHECK-NEXT: ret
|