1 ; RUN: llc < %s -march=arm -mattr=+neon -O0 -regalloc=linearscan
3 ; This test would crash the rewriter when trying to handle a spill after one of
4 ; the @llvm.arm.neon.vld3.v8i8 defined three parts of a register.
6 %struct.__neon_int8x8x3_t = type { <8 x i8>, <8 x i8>, <8 x i8> }
8 declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8*, i32) nounwind readonly
10 declare void @llvm.arm.neon.vst3.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
12 define <8 x i8> @t3(i8* %A1, i8* %A2, i8* %A3, i8* %A4, i8* %A5, i8* %A6, i8* %A7, i8* %A8, i8* %B) nounwind {
13 %tmp1b = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A2, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
14 %tmp2b = extractvalue %struct.__neon_int8x8x3_t %tmp1b, 0 ; <<8 x i8>> [#uses=1]
15 %tmp4b = extractvalue %struct.__neon_int8x8x3_t %tmp1b, 1 ; <<8 x i8>> [#uses=1]
16 %tmp1d = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A4, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
17 %tmp2d = extractvalue %struct.__neon_int8x8x3_t %tmp1d, 0 ; <<8 x i8>> [#uses=1]
18 %tmp4d = extractvalue %struct.__neon_int8x8x3_t %tmp1d, 1 ; <<8 x i8>> [#uses=1]
19 %tmp1e = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A5, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
20 %tmp2e = extractvalue %struct.__neon_int8x8x3_t %tmp1e, 0 ; <<8 x i8>> [#uses=1]
21 %tmp1f = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A6, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
22 %tmp2f = extractvalue %struct.__neon_int8x8x3_t %tmp1f, 0 ; <<8 x i8>> [#uses=1]
23 %tmp1g = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A7, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
24 %tmp2g = extractvalue %struct.__neon_int8x8x3_t %tmp1g, 0 ; <<8 x i8>> [#uses=1]
25 %tmp4g = extractvalue %struct.__neon_int8x8x3_t %tmp1g, 1 ; <<8 x i8>> [#uses=1]
26 %tmp1h = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A8, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
27 %tmp2h = extractvalue %struct.__neon_int8x8x3_t %tmp1h, 0 ; <<8 x i8>> [#uses=1]
28 %tmp3h = extractvalue %struct.__neon_int8x8x3_t %tmp1h, 2 ; <<8 x i8>> [#uses=1]
29 %tmp2bd = add <8 x i8> %tmp2b, %tmp2d ; <<8 x i8>> [#uses=1]
30 %tmp4bd = add <8 x i8> %tmp4b, %tmp4d ; <<8 x i8>> [#uses=1]
31 %tmp2abcd = mul <8 x i8> undef, %tmp2bd ; <<8 x i8>> [#uses=1]
32 %tmp4abcd = mul <8 x i8> undef, %tmp4bd ; <<8 x i8>> [#uses=2]
33 call void @llvm.arm.neon.vst3.v8i8(i8* %A1, <8 x i8> %tmp4abcd, <8 x i8> zeroinitializer, <8 x i8> %tmp2abcd, i32 1)
34 %tmp2ef = sub <8 x i8> %tmp2e, %tmp2f ; <<8 x i8>> [#uses=1]
35 %tmp2gh = sub <8 x i8> %tmp2g, %tmp2h ; <<8 x i8>> [#uses=1]
36 %tmp3gh = sub <8 x i8> zeroinitializer, %tmp3h ; <<8 x i8>> [#uses=1]
37 %tmp4ef = sub <8 x i8> zeroinitializer, %tmp4g ; <<8 x i8>> [#uses=1]
38 %tmp2efgh = mul <8 x i8> %tmp2ef, %tmp2gh ; <<8 x i8>> [#uses=1]
39 %tmp3efgh = mul <8 x i8> undef, %tmp3gh ; <<8 x i8>> [#uses=1]
40 %tmp4efgh = mul <8 x i8> %tmp4ef, undef ; <<8 x i8>> [#uses=2]
41 call void @llvm.arm.neon.vst3.v8i8(i8* %A2, <8 x i8> %tmp4efgh, <8 x i8> %tmp3efgh, <8 x i8> %tmp2efgh, i32 1)
42 %tmp4 = sub <8 x i8> %tmp4efgh, %tmp4abcd ; <<8 x i8>> [#uses=1]
43 tail call void @llvm.arm.neon.vst3.v8i8(i8* %B, <8 x i8> zeroinitializer, <8 x i8> undef, <8 x i8> undef, i32 1)