Lines Matching full:tmp1
6 %tmp1 = load <8 x i8>, <8 x i8>* %A
8 %tmp3 = call <8 x i8> @llvm.aarch64.neon.smax.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
15 %tmp1 = load <16 x i8>, <16 x i8>* %A
17 %tmp3 = call <16 x i8> @llvm.aarch64.neon.smax.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
24 %tmp1 = load <4 x i16>, <4 x i16>* %A
26 %tmp3 = call <4 x i16> @llvm.aarch64.neon.smax.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
33 %tmp1 = load <8 x i16>, <8 x i16>* %A
35 %tmp3 = call <8 x i16> @llvm.aarch64.neon.smax.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
42 %tmp1 = load <2 x i32>, <2 x i32>* %A
44 %tmp3 = call <2 x i32> @llvm.aarch64.neon.smax.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
51 %tmp1 = load <4 x i32>, <4 x i32>* %A
53 %tmp3 = call <4 x i32> @llvm.aarch64.neon.smax.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
67 %tmp1 = load <8 x i8>, <8 x i8>* %A
69 %tmp3 = call <8 x i8> @llvm.aarch64.neon.umax.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
76 %tmp1 = load <16 x i8>, <16 x i8>* %A
78 %tmp3 = call <16 x i8> @llvm.aarch64.neon.umax.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
85 %tmp1 = load <4 x i16>, <4 x i16>* %A
87 %tmp3 = call <4 x i16> @llvm.aarch64.neon.umax.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
94 %tmp1 = load <8 x i16>, <8 x i16>* %A
96 %tmp3 = call <8 x i16> @llvm.aarch64.neon.umax.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
103 %tmp1 = load <2 x i32>, <2 x i32>* %A
105 %tmp3 = call <2 x i32> @llvm.aarch64.neon.umax.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
112 %tmp1 = load <4 x i32>, <4 x i32>* %A
114 %tmp3 = call <4 x i32> @llvm.aarch64.neon.umax.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
128 %tmp1 = load <8 x i8>, <8 x i8>* %A
130 %tmp3 = call <8 x i8> @llvm.aarch64.neon.smin.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
137 %tmp1 = load <16 x i8>, <16 x i8>* %A
139 %tmp3 = call <16 x i8> @llvm.aarch64.neon.smin.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
146 %tmp1 = load <4 x i16>, <4 x i16>* %A
148 %tmp3 = call <4 x i16> @llvm.aarch64.neon.smin.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
155 %tmp1 = load <8 x i16>, <8 x i16>* %A
157 %tmp3 = call <8 x i16> @llvm.aarch64.neon.smin.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
164 %tmp1 = load <2 x i32>, <2 x i32>* %A
166 %tmp3 = call <2 x i32> @llvm.aarch64.neon.smin.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
173 %tmp1 = load <4 x i32>, <4 x i32>* %A
175 %tmp3 = call <4 x i32> @llvm.aarch64.neon.smin.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
189 %tmp1 = load <8 x i8>, <8 x i8>* %A
191 %tmp3 = call <8 x i8> @llvm.aarch64.neon.umin.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
198 %tmp1 = load <16 x i8>, <16 x i8>* %A
200 %tmp3 = call <16 x i8> @llvm.aarch64.neon.umin.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
207 %tmp1 = load <4 x i16>, <4 x i16>* %A
209 %tmp3 = call <4 x i16> @llvm.aarch64.neon.umin.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
216 %tmp1 = load <8 x i16>, <8 x i16>* %A
218 %tmp3 = call <8 x i16> @llvm.aarch64.neon.umin.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
225 %tmp1 = load <2 x i32>, <2 x i32>* %A
227 %tmp3 = call <2 x i32> @llvm.aarch64.neon.umin.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
234 %tmp1 = load <4 x i32>, <4 x i32>* %A
236 %tmp3 = call <4 x i32> @llvm.aarch64.neon.umin.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
252 %tmp1 = load <8 x i8>, <8 x i8>* %A
254 %tmp3 = call <8 x i8> @llvm.aarch64.neon.smaxp.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
261 %tmp1 = load <16 x i8>, <16 x i8>* %A
263 %tmp3 = call <16 x i8> @llvm.aarch64.neon.smaxp.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
270 %tmp1 = load <4 x i16>, <4 x i16>* %A
272 %tmp3 = call <4 x i16> @llvm.aarch64.neon.smaxp.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
279 %tmp1 = load <8 x i16>, <8 x i16>* %A
281 %tmp3 = call <8 x i16> @llvm.aarch64.neon.smaxp.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
288 %tmp1 = load <2 x i32>, <2 x i32>* %A
290 %tmp3 = call <2 x i32> @llvm.aarch64.neon.smaxp.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
297 %tmp1 = load <4 x i32>, <4 x i32>* %A
299 %tmp3 = call <4 x i32> @llvm.aarch64.neon.smaxp.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
313 %tmp1 = load <8 x i8>, <8 x i8>* %A
315 %tmp3 = call <8 x i8> @llvm.aarch64.neon.umaxp.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
322 %tmp1 = load <16 x i8>, <16 x i8>* %A
324 %tmp3 = call <16 x i8> @llvm.aarch64.neon.umaxp.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
331 %tmp1 = load <4 x i16>, <4 x i16>* %A
333 %tmp3 = call <4 x i16> @llvm.aarch64.neon.umaxp.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
340 %tmp1 = load <8 x i16>, <8 x i16>* %A
342 %tmp3 = call <8 x i16> @llvm.aarch64.neon.umaxp.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
349 %tmp1 = load <2 x i32>, <2 x i32>* %A
351 %tmp3 = call <2 x i32> @llvm.aarch64.neon.umaxp.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
358 %tmp1 = load <4 x i32>, <4 x i32>* %A
360 %tmp3 = call <4 x i32> @llvm.aarch64.neon.umaxp.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
376 %tmp1 = load <8 x i8>, <8 x i8>* %A
378 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sminp.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
385 %tmp1 = load <16 x i8>, <16 x i8>* %A
387 %tmp3 = call <16 x i8> @llvm.aarch64.neon.sminp.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
394 %tmp1 = load <4 x i16>, <4 x i16>* %A
396 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sminp.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
403 %tmp1 = load <8 x i16>, <8 x i16>* %A
405 %tmp3 = call <8 x i16> @llvm.aarch64.neon.sminp.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
412 %tmp1 = load <2 x i32>, <2 x i32>* %A
414 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sminp.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
421 %tmp1 = load <4 x i32>, <4 x i32>* %A
423 %tmp3 = call <4 x i32> @llvm.aarch64.neon.sminp.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
437 %tmp1 = load <8 x i8>, <8 x i8>* %A
439 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uminp.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
446 %tmp1 = load <16 x i8>, <16 x i8>* %A
448 %tmp3 = call <16 x i8> @llvm.aarch64.neon.uminp.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
455 %tmp1 = load <4 x i16>, <4 x i16>* %A
457 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uminp.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
464 %tmp1 = load <8 x i16>, <8 x i16>* %A
466 %tmp3 = call <8 x i16> @llvm.aarch64.neon.uminp.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
473 %tmp1 = load <2 x i32>, <2 x i32>* %A
475 %tmp3 = call <2 x i32> @llvm.aarch64.neon.uminp.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
482 %tmp1 = load <4 x i32>, <4 x i32>* %A
484 %tmp3 = call <4 x i32> @llvm.aarch64.neon.uminp.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
498 %tmp1 = load <2 x float>, <2 x float>* %A
500 %tmp3 = call <2 x float> @llvm.aarch64.neon.fmax.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
507 %tmp1 = load <4 x float>, <4 x float>* %A
509 %tmp3 = call <4 x float> @llvm.aarch64.neon.fmax.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
516 %tmp1 = load <2 x double>, <2 x double>* %A
518 %tmp3 = call <2 x double> @llvm.aarch64.neon.fmax.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
529 %tmp1 = load <2 x float>, <2 x float>* %A
531 %tmp3 = call <2 x float> @llvm.aarch64.neon.fmaxp.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
538 %tmp1 = load <4 x float>, <4 x float>* %A
540 %tmp3 = call <4 x float> @llvm.aarch64.neon.fmaxp.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
547 %tmp1 = load <2 x double>, <2 x double>* %A
549 %tmp3 = call <2 x double> @llvm.aarch64.neon.fmaxp.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
560 %tmp1 = load <2 x float>, <2 x float>* %A
562 %tmp3 = call <2 x float> @llvm.aarch64.neon.fmin.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
569 %tmp1 = load <4 x float>, <4 x float>* %A
571 %tmp3 = call <4 x float> @llvm.aarch64.neon.fmin.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
578 %tmp1 = load <2 x double>, <2 x double>* %A
580 %tmp3 = call <2 x double> @llvm.aarch64.neon.fmin.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
591 %tmp1 = load <2 x float>, <2 x float>* %A
593 %tmp3 = call <2 x float> @llvm.aarch64.neon.fminp.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
600 %tmp1 = load <4 x float>, <4 x float>* %A
602 %tmp3 = call <4 x float> @llvm.aarch64.neon.fminp.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
609 %tmp1 = load <2 x double>, <2 x double>* %A
611 %tmp3 = call <2 x double> @llvm.aarch64.neon.fminp.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
622 %tmp1 = load <2 x float>, <2 x float>* %A
624 %tmp3 = call <2 x float> @llvm.aarch64.neon.fminnmp.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
631 %tmp1 = load <4 x float>, <4 x float>* %A
633 %tmp3 = call <4 x float> @llvm.aarch64.neon.fminnmp.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
640 %tmp1 = load <2 x double>, <2 x double>* %A
642 %tmp3 = call <2 x double> @llvm.aarch64.neon.fminnmp.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
653 %tmp1 = load <2 x float>, <2 x float>* %A
655 %tmp3 = call <2 x float> @llvm.aarch64.neon.fmaxnmp.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
662 %tmp1 = load <4 x float>, <4 x float>* %A
664 %tmp3 = call <4 x float> @llvm.aarch64.neon.fmaxnmp.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
671 %tmp1 = load <2 x double>, <2 x double>* %A
673 %tmp3 = call <2 x double> @llvm.aarch64.neon.fmaxnmp.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
