Lines Matching refs:shuffle
50 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
51 %mul = mul <4 x i16> %shuffle, %b
61 …%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32…
62 %mul = mul <8 x i16> %shuffle, %b
72 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
73 %mul = mul <2 x i32> %shuffle, %b
83 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
84 %mul = mul <4 x i32> %shuffle, %b
94 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
95 %mul = mul <4 x i16> %shuffle, %b
105 …%shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> <i32 7, i32 7, i32 7, i32 7, i32…
106 %mul = mul <8 x i16> %shuffle, %b
116 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
117 %mul = mul <2 x i32> %shuffle, %b
127 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
128 %mul = mul <4 x i32> %shuffle, %b
138 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
139 %mul = mul <4 x i16> %shuffle, %b
149 …%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32…
150 %mul = mul <8 x i16> %shuffle, %b
160 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
161 %mul = mul <2 x i32> %shuffle, %b
171 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
172 %mul = mul <4 x i32> %shuffle, %b
182 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
183 %mul = mul <4 x i16> %shuffle, %b
193 …%shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> <i32 7, i32 7, i32 7, i32 7, i32…
194 %mul = mul <8 x i16> %shuffle, %b
204 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
205 %mul = mul <2 x i32> %shuffle, %b
215 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
216 %mul = mul <4 x i32> %shuffle, %b
226 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
227 %mul = mul <4 x i16> %shuffle, %a
236 …%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32…
237 %mul = mul <8 x i16> %shuffle, %a
246 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
247 %mul = mul <2 x i32> %shuffle, %a
256 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
257 %mul = mul <4 x i32> %shuffle, %a
266 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
267 %mul = mul <4 x i16> %shuffle, %a
276 …%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32…
277 %mul = mul <8 x i16> %shuffle, %a
286 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
287 %mul = mul <2 x i32> %shuffle, %a
296 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
297 %mul = mul <4 x i32> %shuffle, %a
306 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
307 %mul = mul <4 x i16> %shuffle, %a
316 …%shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> <i32 7, i32 7, i32 7, i32 7, i32…
317 %mul = mul <8 x i16> %shuffle, %a
326 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
327 %mul = mul <2 x i32> %shuffle, %a
336 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
337 %mul = mul <4 x i32> %shuffle, %a
346 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
347 %mul = mul <4 x i16> %shuffle, %a
356 …%shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> <i32 7, i32 7, i32 7, i32 7, i32…
357 %mul = mul <8 x i16> %shuffle, %a
366 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
367 %mul = mul <2 x i32> %shuffle, %a
376 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
377 %mul = mul <4 x i32> %shuffle, %a
620 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
621 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
631 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
632 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
642 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
643 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
653 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
654 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
664 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
665 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
666 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
676 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
677 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
678 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
688 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
689 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
690 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
700 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
701 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
702 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
712 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
713 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
723 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
724 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
734 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
735 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
745 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
746 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
756 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
757 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
758 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
768 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
769 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
770 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
780 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
781 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
782 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
792 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
793 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
794 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
804 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
805 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
815 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
816 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
826 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
827 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
837 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
838 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
848 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
849 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
850 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
860 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
861 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
862 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
872 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
873 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
874 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
884 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
885 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
886 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
896 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
897 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
907 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
908 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
918 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
919 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
929 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
930 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
940 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
941 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
942 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
952 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
953 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
954 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
964 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
965 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
966 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
976 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
977 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
978 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
988 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
989 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
998 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
999 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
1008 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1009 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
1018 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1019 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
1028 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1029 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1030 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1039 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1040 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1041 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1050 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1051 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1052 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1061 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1062 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1063 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1072 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
1073 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
1082 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
1083 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
1092 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
1093 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
1102 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
1103 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
1112 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1113 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
1114 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1123 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1124 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
1125 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1134 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1135 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
1136 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1145 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1146 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
1147 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1156 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1157 …vqdmlal2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
1167 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1168 …vqdmlal2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
1178 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1179 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1180 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1190 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1191 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1192 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1202 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1203 …vqdmlsl2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
1213 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1214 …vqdmlsl2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
1224 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1225 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1226 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1236 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1237 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1238 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1248 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1249 …vqdmull2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
1258 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1259 …vqdmull2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
1268 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1269 …vqdmull2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
1278 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
1279 …vqdmull2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
1288 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1289 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1290 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1299 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1300 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1301 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1310 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
1311 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> <i32 7, i32 7, i32 7, i32 7>
1312 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
1321 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
1322 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> <i32 3, i32 3>
1323 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
1332 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1333 …vqdmulh2.i = tail call <4 x i16> @llvm.aarch64.neon.sqdmulh.v4i16(<4 x i16> %a, <4 x i16> %shuffle)
1342 …%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32…
1343 …vqdmulh2.i = tail call <8 x i16> @llvm.aarch64.neon.sqdmulh.v8i16(<8 x i16> %a, <8 x i16> %shuffle)
1352 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1353 …vqdmulh2.i = tail call <2 x i32> @llvm.aarch64.neon.sqdmulh.v2i32(<2 x i32> %a, <2 x i32> %shuffle)
1362 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
1363 …vqdmulh2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmulh.v4i32(<4 x i32> %a, <4 x i32> %shuffle)
1372 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1373 …rdmulh2.i = tail call <4 x i16> @llvm.aarch64.neon.sqrdmulh.v4i16(<4 x i16> %a, <4 x i16> %shuffle)
1382 …%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32…
1383 …rdmulh2.i = tail call <8 x i16> @llvm.aarch64.neon.sqrdmulh.v8i16(<8 x i16> %a, <8 x i16> %shuffle)
1392 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
1393 …rdmulh2.i = tail call <2 x i32> @llvm.aarch64.neon.sqrdmulh.v2i32(<2 x i32> %a, <2 x i32> %shuffle)
1402 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
1403 …rdmulh2.i = tail call <4 x i32> @llvm.aarch64.neon.sqrdmulh.v4i32(<4 x i32> %a, <4 x i32> %shuffle)
1412 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <2 x i32> <i32 1, i32 1>
1413 %mul = fmul <2 x float> %shuffle, %a
1435 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
1436 %mul = fmul <4 x float> %shuffle, %a
1445 %shuffle = shufflevector <1 x double> %v, <1 x double> undef, <2 x i32> zeroinitializer
1446 %mul = fmul <2 x double> %shuffle, %a
1455 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <2 x i32> <i32 3, i32 3>
1456 %mul = fmul <2 x float> %shuffle, %a
1478 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1479 %mul = fmul <4 x float> %shuffle, %a
1488 %shuffle = shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> <i32 1, i32 1>
1489 %mul = fmul <2 x double> %shuffle, %a
1498 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <2 x i32> <i32 1, i32 1>
1499 …ulx2.i = tail call <2 x float> @llvm.aarch64.neon.fmulx.v2f32(<2 x float> %a, <2 x float> %shuffle)
1508 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
1509 …ulx2.i = tail call <4 x float> @llvm.aarch64.neon.fmulx.v4f32(<4 x float> %a, <4 x float> %shuffle)
1518 %shuffle = shufflevector <1 x double> %v, <1 x double> undef, <2 x i32> zeroinitializer
1519 …2.i = tail call <2 x double> @llvm.aarch64.neon.fmulx.v2f64(<2 x double> %a, <2 x double> %shuffle)
1528 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <2 x i32> <i32 3, i32 3>
1529 …ulx2.i = tail call <2 x float> @llvm.aarch64.neon.fmulx.v2f32(<2 x float> %a, <2 x float> %shuffle)
1538 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
1539 …ulx2.i = tail call <4 x float> @llvm.aarch64.neon.fmulx.v4f32(<4 x float> %a, <4 x float> %shuffle)
1548 %shuffle = shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> <i32 1, i32 1>
1549 …2.i = tail call <2 x double> @llvm.aarch64.neon.fmulx.v2f64(<2 x double> %a, <2 x double> %shuffle)
1558 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
1559 %mul = mul <4 x i16> %shuffle, %b
1569 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> zeroinitializer
1570 %mul = mul <8 x i16> %shuffle, %b
1580 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
1581 %mul = mul <2 x i32> %shuffle, %b
1591 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> zeroinitializer
1592 %mul = mul <4 x i32> %shuffle, %b
1602 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
1603 %mul = mul <4 x i16> %shuffle, %b
1613 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> zeroinitializer
1614 %mul = mul <8 x i16> %shuffle, %b
1624 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
1625 %mul = mul <2 x i32> %shuffle, %b
1635 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> zeroinitializer
1636 %mul = mul <4 x i32> %shuffle, %b
1646 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
1647 %mul = mul <4 x i16> %shuffle, %b
1657 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> zeroinitializer
1658 %mul = mul <8 x i16> %shuffle, %b
1668 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
1669 %mul = mul <2 x i32> %shuffle, %b
1679 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> zeroinitializer
1680 %mul = mul <4 x i32> %shuffle, %b
1690 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
1691 %mul = mul <4 x i16> %shuffle, %b
1701 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> zeroinitializer
1702 %mul = mul <8 x i16> %shuffle, %b
1712 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
1713 %mul = mul <2 x i32> %shuffle, %b
1723 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> zeroinitializer
1724 %mul = mul <4 x i32> %shuffle, %b
1734 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
1735 %mul = mul <4 x i16> %shuffle, %a
1744 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> zeroinitializer
1745 %mul = mul <8 x i16> %shuffle, %a
1754 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
1755 %mul = mul <2 x i32> %shuffle, %a
1764 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> zeroinitializer
1765 %mul = mul <4 x i32> %shuffle, %a
1774 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
1775 %mul = mul <4 x i16> %shuffle, %a
1784 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> zeroinitializer
1785 %mul = mul <8 x i16> %shuffle, %a
1794 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
1795 %mul = mul <2 x i32> %shuffle, %a
1804 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> zeroinitializer
1805 %mul = mul <4 x i32> %shuffle, %a
1814 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
1815 %mul = mul <4 x i16> %shuffle, %a
1824 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> zeroinitializer
1825 %mul = mul <8 x i16> %shuffle, %a
1834 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
1835 %mul = mul <2 x i32> %shuffle, %a
1844 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> zeroinitializer
1845 %mul = mul <4 x i32> %shuffle, %a
1854 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
1855 %mul = mul <4 x i16> %shuffle, %a
1864 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <8 x i32> zeroinitializer
1865 %mul = mul <8 x i16> %shuffle, %a
1874 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
1875 %mul = mul <2 x i32> %shuffle, %a
1884 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <4 x i32> zeroinitializer
1885 %mul = mul <4 x i32> %shuffle, %a
1999 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2000 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2010 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2011 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2021 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2022 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2032 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2033 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2043 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2044 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2045 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2055 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2056 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2057 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2067 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2068 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2069 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2079 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2080 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2081 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2091 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2092 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2102 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2103 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2113 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2114 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2124 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2125 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2135 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2136 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2137 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2147 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2148 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2149 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2159 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2160 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2161 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2171 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2172 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2173 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2183 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2184 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2194 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2195 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2205 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2206 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2216 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2217 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2227 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2228 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2229 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2239 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2240 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2241 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2251 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2252 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2253 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2263 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2264 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2265 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2275 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2276 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2286 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2287 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2297 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2298 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2308 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2309 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2319 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2320 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2321 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2331 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2332 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2333 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2343 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2344 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2345 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2355 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2356 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2357 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2367 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2368 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
2377 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2378 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
2387 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2388 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
2397 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2398 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
2407 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2408 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2409 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2418 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2419 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2420 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2429 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2430 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2431 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2440 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2441 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2442 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2451 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2452 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
2461 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2462 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
2471 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2472 %vmull2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
2481 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2482 %vmull2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
2491 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2492 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2493 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.smull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2502 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2503 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2504 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.smull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2513 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2514 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2515 …l2.i = tail call <4 x i32> @llvm.aarch64.neon.umull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2524 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2525 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2526 …l2.i = tail call <2 x i64> @llvm.aarch64.neon.umull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2535 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2536 …vqdmlal2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2546 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2547 …vqdmlal2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2557 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2558 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2559 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2569 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2570 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2571 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2581 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2582 …vqdmlsl2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %b, <4 x i16> %shuffle)
2592 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2593 …vqdmlsl2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %b, <2 x i32> %shuffle)
2603 %shuffle.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2604 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2605 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2615 %shuffle.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2616 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2617 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2627 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2628 …vqdmull2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
2637 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2638 …vqdmull2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
2647 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2648 …vqdmull2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %a, <4 x i16> %shuffle)
2657 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2658 …vqdmull2.i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %a, <2 x i32> %shuffle)
2667 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2668 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2669 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2678 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2679 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2680 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2689 %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
2690 %shuffle = shufflevector <8 x i16> %v, <8 x i16> undef, <4 x i32> zeroinitializer
2691 ….i = tail call <4 x i32> @llvm.aarch64.neon.sqdmull.v4i32(<4 x i16> %shuffle.i, <4 x i16> %shuffle)
2700 %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
2701 %shuffle = shufflevector <4 x i32> %v, <4 x i32> undef, <2 x i32> zeroinitializer
2702 ….i = tail call <2 x i64> @llvm.aarch64.neon.sqdmull.v2i64(<2 x i32> %shuffle.i, <2 x i32> %shuffle)
2711 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2712 …vqdmulh2.i = tail call <4 x i16> @llvm.aarch64.neon.sqdmulh.v4i16(<4 x i16> %a, <4 x i16> %shuffle)
2721 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> zeroinitializer
2722 …vqdmulh2.i = tail call <8 x i16> @llvm.aarch64.neon.sqdmulh.v8i16(<8 x i16> %a, <8 x i16> %shuffle)
2731 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2732 …vqdmulh2.i = tail call <2 x i32> @llvm.aarch64.neon.sqdmulh.v2i32(<2 x i32> %a, <2 x i32> %shuffle)
2741 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> zeroinitializer
2742 …vqdmulh2.i = tail call <4 x i32> @llvm.aarch64.neon.sqdmulh.v4i32(<4 x i32> %a, <4 x i32> %shuffle)
2751 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> zeroinitializer
2752 …rdmulh2.i = tail call <4 x i16> @llvm.aarch64.neon.sqrdmulh.v4i16(<4 x i16> %a, <4 x i16> %shuffle)
2761 %shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <8 x i32> zeroinitializer
2762 …rdmulh2.i = tail call <8 x i16> @llvm.aarch64.neon.sqrdmulh.v8i16(<8 x i16> %a, <8 x i16> %shuffle)
2771 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> zeroinitializer
2772 …rdmulh2.i = tail call <2 x i32> @llvm.aarch64.neon.sqrdmulh.v2i32(<2 x i32> %a, <2 x i32> %shuffle)
2781 %shuffle = shufflevector <2 x i32> %v, <2 x i32> undef, <4 x i32> zeroinitializer
2782 …rdmulh2.i = tail call <4 x i32> @llvm.aarch64.neon.sqrdmulh.v4i32(<4 x i32> %a, <4 x i32> %shuffle)
2791 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <2 x i32> zeroinitializer
2792 %mul = fmul <2 x float> %shuffle, %a
2801 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <4 x i32> zeroinitializer
2802 %mul = fmul <4 x float> %shuffle, %a
2811 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <2 x i32> zeroinitializer
2812 %mul = fmul <2 x float> %shuffle, %a
2834 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <4 x i32> zeroinitializer
2835 %mul = fmul <4 x float> %shuffle, %a
2844 %shuffle = shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> zeroinitializer
2845 %mul = fmul <2 x double> %shuffle, %a
2854 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <2 x i32> zeroinitializer
2855 …ulx2.i = tail call <2 x float> @llvm.aarch64.neon.fmulx.v2f32(<2 x float> %a, <2 x float> %shuffle)
2864 %shuffle = shufflevector <2 x float> %v, <2 x float> undef, <4 x i32> zeroinitializer
2865 …ulx2.i = tail call <4 x float> @llvm.aarch64.neon.fmulx.v4f32(<4 x float> %a, <4 x float> %shuffle)
2874 %shuffle = shufflevector <1 x double> %v, <1 x double> undef, <2 x i32> zeroinitializer
2875 …2.i = tail call <2 x double> @llvm.aarch64.neon.fmulx.v2f64(<2 x double> %a, <2 x double> %shuffle)
2884 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <2 x i32> zeroinitializer
2885 …ulx2.i = tail call <2 x float> @llvm.aarch64.neon.fmulx.v2f32(<2 x float> %a, <2 x float> %shuffle)
2894 %shuffle = shufflevector <4 x float> %v, <4 x float> undef, <4 x i32> zeroinitializer
2895 …ulx2.i = tail call <4 x float> @llvm.aarch64.neon.fmulx.v4f32(<4 x float> %a, <4 x float> %shuffle)
2904 %shuffle = shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> zeroinitializer
2905 …2.i = tail call <2 x double> @llvm.aarch64.neon.fmulx.v2f64(<2 x double> %a, <2 x double> %shuffle)