1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
|
; RUN: llc -mtriple=x86_64-apple-darwin -mattr=+sse2 < %s | FileCheck --check-prefix=SSE2-CODEGEN %s
; RUN: opt -mtriple=x86_64-apple-darwin -mattr=+sse2 -passes="print<cost-model>" 2>&1 -disable-output < %s | FileCheck --check-prefix=SSE2 %s
define <4 x i32> @shl(<4 x i32> %vector, i32 %scalar) {
entry:
; SSE2: 'shl'
; SSE2: cost of 2 {{.*}} shl
; SSE2-CODEGEN: movd %edi, %xmm1
; SSE2-CODEGEN: pslld %xmm1, %xmm0
%insert = insertelement <4 x i32> poison, i32 %scalar, i32 0
%splat = shufflevector <4 x i32> %insert, <4 x i32> poison, <4 x i32> zeroinitializer
%ret = shl <4 x i32> %vector , %splat
ret <4 x i32> %ret
}
define <4 x i32> @ashr(<4 x i32> %vector, i32 %scalar) {
entry:
; SSE2: 'ashr'
; SSE2: cost of 2 {{.*}} ashr
; SSE2-CODEGEN: movd %edi, %xmm1
; SSE2-CODEGEN: psrad %xmm1, %xmm0
%insert = insertelement <4 x i32> poison, i32 %scalar, i32 0
%splat = shufflevector <4 x i32> %insert, <4 x i32> poison, <4 x i32> zeroinitializer
%ret = ashr <4 x i32> %vector , %splat
ret <4 x i32> %ret
}
define <4 x i32> @lshr(<4 x i32> %vector, i32 %scalar) {
entry:
; SSE2: 'lshr'
; SSE2: cost of 2 {{.*}} lshr
; SSE2-CODEGEN: movd %edi, %xmm1
; SSE2-CODEGEN: psrld %xmm1, %xmm0
%insert = insertelement <4 x i32> poison, i32 %scalar, i32 0
%splat = shufflevector <4 x i32> %insert, <4 x i32> poison, <4 x i32> zeroinitializer
%ret = lshr <4 x i32> %vector , %splat
ret <4 x i32> %ret
}
|
