1; RUN: llc < %s -march=x86 -mattr=+sse2 -stats |& grep {Number of 3-address instructions sunk} 2 3define void @t2(<2 x i64>* %vDct, <2 x i64>* %vYp, i8* %skiplist, <2 x i64> %a1) nounwind { 4entry: 5 %tmp25 = bitcast <2 x i64> %a1 to <8 x i16> ; <<8 x i16>> [#uses=1] 6 br label %bb 7bb: ; preds = %bb, %entry 8 %skiplist_addr.0.rec = phi i32 [ 0, %entry ], [ %indvar.next, %bb ] ; <i32> [#uses=3] 9 %vYp_addr.0.rec = shl i32 %skiplist_addr.0.rec, 3 ; <i32> [#uses=3] 10 %vDct_addr.0 = getelementptr <2 x i64>* %vDct, i32 %vYp_addr.0.rec ; <<2 x i64>*> [#uses=1] 11 %vYp_addr.0 = getelementptr <2 x i64>* %vYp, i32 %vYp_addr.0.rec ; <<2 x i64>*> [#uses=1] 12 %skiplist_addr.0 = getelementptr i8* %skiplist, i32 %skiplist_addr.0.rec ; <i8*> [#uses=1] 13 %vDct_addr.0.sum43 = or i32 %vYp_addr.0.rec, 1 ; <i32> [#uses=1] 14 %tmp7 = getelementptr <2 x i64>* %vDct, i32 %vDct_addr.0.sum43 ; <<2 x i64>*> [#uses=1] 15 %tmp8 = load <2 x i64>* %tmp7, align 16 ; <<2 x i64>> [#uses=1] 16 %tmp11 = load <2 x i64>* %vDct_addr.0, align 16 ; <<2 x i64>> [#uses=1] 17 %tmp13 = bitcast <2 x i64> %tmp8 to <8 x i16> ; <<8 x i16>> [#uses=1] 18 %tmp15 = bitcast <2 x i64> %tmp11 to <8 x i16> ; <<8 x i16>> [#uses=1] 19 %tmp16 = shufflevector <8 x i16> %tmp15, <8 x i16> %tmp13, <8 x i32> < i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11 > ; <<8 x i16>> [#uses=1] 20 %tmp26 = mul <8 x i16> %tmp25, %tmp16 ; <<8 x i16>> [#uses=1] 21 %tmp27 = bitcast <8 x i16> %tmp26 to <2 x i64> ; <<2 x i64>> [#uses=1] 22 store <2 x i64> %tmp27, <2 x i64>* %vYp_addr.0, align 16 23 %tmp37 = load i8* %skiplist_addr.0, align 1 ; <i8> [#uses=1] 24 %tmp38 = icmp eq i8 %tmp37, 0 ; <i1> [#uses=1] 25 %indvar.next = add i32 %skiplist_addr.0.rec, 1 ; <i32> [#uses=1] 26 br i1 %tmp38, label %return, label %bb 27return: ; preds = %bb 28 ret void 29} 30