1; RUN: llc < %s -march=x86-64 -mcpu=corei7 | FileCheck %s 2 3target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i8:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128" 4target triple = "x86_64-unknown-linux-gnu" 5 6 7; CHECK: mul_f 8define i32 @mul_f(<4 x i8>* %A) { 9entry: 10; CHECK: pmul 11; CHECK-NOT: mulb 12 %0 = load <4 x i8>, <4 x i8>* %A, align 8 13 %mul = mul <4 x i8> %0, %0 14 store <4 x i8> %mul, <4 x i8>* undef 15 ret i32 0 16; CHECK: ret 17} 18 19 20; CHECK: shuff_f 21define i32 @shuff_f(<4 x i8>* %A) { 22entry: 23; CHECK: pmovzxbd 24; CHECK: paddd 25; CHECK: pshufb 26 %0 = load <4 x i8>, <4 x i8>* %A, align 8 27 %add = add <4 x i8> %0, %0 28 store <4 x i8> %add, <4 x i8>* undef 29 ret i32 0 30; CHECK: ret 31} 32 33; CHECK: bitcast_widen 34define <2 x float> @bitcast_widen(<4 x i32> %in) nounwind readnone { 35entry: 36; CHECK-NOT: pshufd 37 %x = shufflevector <4 x i32> %in, <4 x i32> undef, <2 x i32> <i32 0, i32 1> 38 %y = bitcast <2 x i32> %x to <2 x float> 39 ret <2 x float> %y 40; CHECK: ret 41} 42 43