File: vec_extract-sse4.ll

package info (click to toggle)
llvm 2.6-9.1
  • links: PTS
  • area: main
  • in suites: squeeze
  • size: 57,604 kB
  • ctags: 44,336
  • sloc: cpp: 344,766; sh: 12,407; ansic: 10,617; ada: 3,070; ml: 2,505; perl: 2,496; makefile: 1,426; pascal: 1,163; exp: 389; asm: 307; python: 298; objc: 260; lisp: 182; csh: 117; xml: 38; f90: 36; tcl: 20
file content (31 lines) | stat: -rw-r--r-- 784 bytes parent folder | download 
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
 
; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse41 -o %t -f
; RUN: grep extractps   %t | count 1
; RUN: grep pextrd      %t | count 1
; RUN: not grep pshufd  %t
; RUN: not grep movss   %t

define void @t1(float* %R, <4 x float>* %P1) nounwind {
	%X = load <4 x float>* %P1
	%tmp = extractelement <4 x float> %X, i32 3
	store float %tmp, float* %R
	ret void
}

define float @t2(<4 x float>* %P1) nounwind {
	%X = load <4 x float>* %P1
	%tmp = extractelement <4 x float> %X, i32 2
	ret float %tmp
}

define void @t3(i32* %R, <4 x i32>* %P1) nounwind {
	%X = load <4 x i32>* %P1
	%tmp = extractelement <4 x i32> %X, i32 3
	store i32 %tmp, i32* %R
	ret void
}

define i32 @t4(<4 x i32>* %P1) nounwind {
	%X = load <4 x i32>* %P1
	%tmp = extractelement <4 x i32> %X, i32 3
	ret i32 %tmp
}