File: vec_shift2.ll

package info (click to toggle)
llvm-2.7 2.7-6
  • links: PTS
  • area: main
  • in suites: squeeze
  • size: 59,992 kB
  • ctags: 49,499
  • sloc: cpp: 373,792; ansic: 16,885; sh: 12,614; asm: 6,809; ada: 3,083; ml: 2,942; python: 2,671; perl: 2,404; makefile: 1,691; pascal: 1,235; exp: 399; objc: 291; lisp: 184; csh: 117; xml: 38; f90: 36
file content (17 lines) | stat: -rw-r--r-- 843 bytes parent folder | download | duplicates (11) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
 
; RUN: llc < %s -march=x86 -mattr=+sse2 | not grep CPI

define <2 x i64> @t1(<2 x i64> %b1, <2 x i64> %c) nounwind  {
	%tmp1 = bitcast <2 x i64> %b1 to <8 x i16>
	%tmp2 = tail call <8 x i16> @llvm.x86.sse2.psrl.w( <8 x i16> %tmp1, <8 x i16> bitcast (<4 x i32> < i32 14, i32 undef, i32 undef, i32 undef > to <8 x i16>) ) nounwind readnone
	%tmp3 = bitcast <8 x i16> %tmp2 to <2 x i64>
	ret <2 x i64> %tmp3
}

define <4 x i32> @t2(<2 x i64> %b1, <2 x i64> %c) nounwind  {
	%tmp1 = bitcast <2 x i64> %b1 to <4 x i32>
	%tmp2 = tail call <4 x i32> @llvm.x86.sse2.psll.d( <4 x i32> %tmp1, <4 x i32> < i32 14, i32 undef, i32 undef, i32 undef > ) nounwind readnone
	ret <4 x i32> %tmp2
}

declare <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16>, <8 x i16>) nounwind readnone 
declare <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32>, <4 x i32>) nounwind readnone