mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-31 07:52:55 +01:00
e4b5559cf8
instead of to have a chained series of scope nodes. This makes the generated table smaller, improves the efficiency of the interpreter, and make the factoring optimization much more reasonable to implement. llvm-svn: 97160
16 lines
608 B
LLVM
16 lines
608 B
LLVM
; RUN: llc < %s -march=x86 -mattr=sse | grep movaps
|
|
; Test that the load is NOT folded into the intrinsic, which would zero the top
|
|
; elts of the loaded vector.
|
|
|
|
target datalayout = "e-p:32:32"
|
|
target triple = "i686-apple-darwin8.7.2"
|
|
|
|
define <4 x float> @test(<4 x float> %A, <4 x float>* %B) nounwind {
|
|
%BV = load <4 x float>* %B ; <<4 x float>> [#uses=1]
|
|
%tmp28 = tail call <4 x float> @llvm.x86.sse.sub.ss( <4 x float> %A, <4 x float> %BV ) ; <<4 x float>> [#uses=1]
|
|
ret <4 x float> %tmp28
|
|
}
|
|
|
|
declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>)
|
|
|