Lines Matching full:neon
1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
9 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
19 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
29 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
41 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
53 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
65 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
75 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
85 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
95 %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
108 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
120 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
132 %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
142 %tmp3 = call <2 x float> @llvm.aarch64.neon.fabd.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
151 %tmp3 = call <4 x float> @llvm.aarch64.neon.fabd.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
160 %tmp3 = call <2 x double> @llvm.aarch64.neon.fabd.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
164 declare <2 x float> @llvm.aarch64.neon.fabd.v2f32(<2 x float>, <2 x float>) nounwind readnone
165 declare <4 x float> @llvm.aarch64.neon.fabd.v4f32(<4 x float>, <4 x float>) nounwind readnone
166 declare <2 x double> @llvm.aarch64.neon.fabd.v2f64(<2 x double>, <2 x double>) nounwind readnone
173 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
182 %tmp3 = call <16 x i8> @llvm.aarch64.neon.sabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
191 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
200 %tmp3 = call <8 x i16> @llvm.aarch64.neon.sabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
209 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
218 %tmp3 = call <4 x i32> @llvm.aarch64.neon.sabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
222 declare <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
223 declare <16 x i8> @llvm.aarch64.neon.sabd.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
224 declare <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
225 declare <8 x i16> @llvm.aarch64.neon.sabd.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
226 declare <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32>, <2 x i32>) nounwind readnone
227 declare <4 x i32> @llvm.aarch64.neon.sabd.v4i32(<4 x i32>, <4 x i32>) nounwind readnone
234 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
243 %tmp3 = call <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
252 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
261 %tmp3 = call <8 x i16> @llvm.aarch64.neon.uabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
270 %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
279 %tmp3 = call <4 x i32> @llvm.aarch64.neon.uabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
283 declare <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
284 declare <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
285 declare <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
286 declare <8 x i16> @llvm.aarch64.neon.uabd.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
287 declare <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32>, <2 x i32>) nounwind readnone
288 declare <4 x i32> @llvm.aarch64.neon.uabd.v4i32(<4 x i32>, <4 x i32>) nounwind readnone
294 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sqabs.v8i8(<8 x i8> %tmp1)
302 %tmp3 = call <16 x i8> @llvm.aarch64.neon.sqabs.v16i8(<16 x i8> %tmp1)
310 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sqabs.v4i16(<4 x i16> %tmp1)
318 %tmp3 = call <8 x i16> @llvm.aarch64.neon.sqabs.v8i16(<8 x i16> %tmp1)
326 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sqabs.v2i32(<2 x i32> %tmp1)
334 %tmp3 = call <4 x i32> @llvm.aarch64.neon.sqabs.v4i32(<4 x i32> %tmp1)
338 declare <8 x i8> @llvm.aarch64.neon.sqabs.v8i8(<8 x i8>) nounwind readnone
339 declare <16 x i8> @llvm.aarch64.neon.sqabs.v16i8(<16 x i8>) nounwind readnone
340 declare <4 x i16> @llvm.aarch64.neon.sqabs.v4i16(<4 x i16>) nounwind readnone
341 declare <8 x i16> @llvm.aarch64.neon.sqabs.v8i16(<8 x i16>) nounwind readnone
342 declare <2 x i32> @llvm.aarch64.neon.sqabs.v2i32(<2 x i32>) nounwind readnone
343 declare <4 x i32> @llvm.aarch64.neon.sqabs.v4i32(<4 x i32>) nounwind readnone
349 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sqneg.v8i8(<8 x i8> %tmp1)
357 %tmp3 = call <16 x i8> @llvm.aarch64.neon.sqneg.v16i8(<16 x i8> %tmp1)
365 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sqneg.v4i16(<4 x i16> %tmp1)
373 %tmp3 = call <8 x i16> @llvm.aarch64.neon.sqneg.v8i16(<8 x i16> %tmp1)
381 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sqneg.v2i32(<2 x i32> %tmp1)
389 %tmp3 = call <4 x i32> @llvm.aarch64.neon.sqneg.v4i32(<4 x i32> %tmp1)
393 declare <8 x i8> @llvm.aarch64.neon.sqneg.v8i8(<8 x i8>) nounwind readnone
394 declare <16 x i8> @llvm.aarch64.neon.sqneg.v16i8(<16 x i8>) nounwind readnone
395 declare <4 x i16> @llvm.aarch64.neon.sqneg.v4i16(<4 x i16>) nounwind readnone
396 declare <8 x i16> @llvm.aarch64.neon.sqneg.v8i16(<8 x i16>) nounwind readnone
397 declare <2 x i32> @llvm.aarch64.neon.sqneg.v2i32(<2 x i32>) nounwind readnone
398 declare <4 x i32> @llvm.aarch64.neon.sqneg.v4i32(<4 x i32>) nounwind readnone
404 %tmp3 = call <8 x i8> @llvm.aarch64.neon.abs.v8i8(<8 x i8> %tmp1)
412 %tmp3 = call <16 x i8> @llvm.aarch64.neon.abs.v16i8(<16 x i8> %tmp1)
420 %tmp3 = call <4 x i16> @llvm.aarch64.neon.abs.v4i16(<4 x i16> %tmp1)
428 %tmp3 = call <8 x i16> @llvm.aarch64.neon.abs.v8i16(<8 x i16> %tmp1)
436 %tmp3 = call <2 x i32> @llvm.aarch64.neon.abs.v2i32(<2 x i32> %tmp1)
444 %tmp3 = call <4 x i32> @llvm.aarch64.neon.abs.v4i32(<4 x i32> %tmp1)
451 %abs = call <1 x i64> @llvm.aarch64.neon.abs.v1i64(<1 x i64> %A)
458 %abs = call i64 @llvm.aarch64.neon.abs.i64(i64 %A)
462 declare <8 x i8> @llvm.aarch64.neon.abs.v8i8(<8 x i8>) nounwind readnone
463 declare <16 x i8> @llvm.aarch64.neon.abs.v16i8(<16 x i8>) nounwind readnone
464 declare <4 x i16> @llvm.aarch64.neon.abs.v4i16(<4 x i16>) nounwind readnone
465 declare <8 x i16> @llvm.aarch64.neon.abs.v8i16(<8 x i16>) nounwind readnone
466 declare <2 x i32> @llvm.aarch64.neon.abs.v2i32(<2 x i32>) nounwind readnone
467 declare <4 x i32> @llvm.aarch64.neon.abs.v4i32(<4 x i32>) nounwind readnone
468 declare <1 x i64> @llvm.aarch64.neon.abs.v1i64(<1 x i64>) nounwind readnone
469 declare i64 @llvm.aarch64.neon.abs.i64(i64) nounwind readnone
477 %tmp4 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
489 %tmp4 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
501 %tmp4 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
516 %tmp4 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
530 %tmp4 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
544 %tmp4 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
556 %tmp4 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
568 %tmp4 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
580 %tmp4 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
594 %tmp4 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
608 %tmp4 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
622 %tmp4 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
633 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
644 %tmp3 = call <16 x i8> @llvm.aarch64.neon.sabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
655 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
666 %tmp3 = call <8 x i16> @llvm.aarch64.neon.sabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
677 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
688 %tmp3 = call <4 x i32> @llvm.aarch64.neon.sabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
699 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
710 %tmp3 = call <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
721 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
732 %tmp3 = call <8 x i16> @llvm.aarch64.neon.uabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
743 %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
754 %tmp3 = call <4 x i32> @llvm.aarch64.neon.uabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
787 %res = tail call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %lhs.high, <2 x i32> %rhsvec) nounwind
801 %res = tail call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %lhs.high, <2 x i32> %rhsvec) nounwind
