1 ; RUN: llc -march=mips -mattr=+msa < %s | FileCheck %s
3 @llvm_mips_bclr_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
4 @llvm_mips_bclr_b_ARG2 = global <16 x i8> <i8 16, i8 17, i8 18, i8 19, i8 20, i8 21, i8 22, i8 23, i8 24, i8 25, i8 26, i8 27, i8 28, i8 29, i8 30, i8 31>, align 16
5 @llvm_mips_bclr_b_RES = global <16 x i8> <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, align 16
7 define void @llvm_mips_bclr_b_test() nounwind {
9 %0 = load <16 x i8>* @llvm_mips_bclr_b_ARG1
10 %1 = load <16 x i8>* @llvm_mips_bclr_b_ARG2
11 %2 = tail call <16 x i8> @llvm.mips.bclr.b(<16 x i8> %0, <16 x i8> %1)
12 store <16 x i8> %2, <16 x i8>* @llvm_mips_bclr_b_RES
16 declare <16 x i8> @llvm.mips.bclr.b(<16 x i8>, <16 x i8>) nounwind
18 ; CHECK: llvm_mips_bclr_b_test:
23 ; CHECK: .size llvm_mips_bclr_b_test
25 @llvm_mips_bclr_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
26 @llvm_mips_bclr_h_ARG2 = global <8 x i16> <i16 8, i16 9, i16 10, i16 11, i16 12, i16 13, i16 14, i16 15>, align 16
27 @llvm_mips_bclr_h_RES = global <8 x i16> <i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>, align 16
29 define void @llvm_mips_bclr_h_test() nounwind {
31 %0 = load <8 x i16>* @llvm_mips_bclr_h_ARG1
32 %1 = load <8 x i16>* @llvm_mips_bclr_h_ARG2
33 %2 = tail call <8 x i16> @llvm.mips.bclr.h(<8 x i16> %0, <8 x i16> %1)
34 store <8 x i16> %2, <8 x i16>* @llvm_mips_bclr_h_RES
38 declare <8 x i16> @llvm.mips.bclr.h(<8 x i16>, <8 x i16>) nounwind
40 ; CHECK: llvm_mips_bclr_h_test:
45 ; CHECK: .size llvm_mips_bclr_h_test
47 @llvm_mips_bclr_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
48 @llvm_mips_bclr_w_ARG2 = global <4 x i32> <i32 4, i32 5, i32 6, i32 7>, align 16
49 @llvm_mips_bclr_w_RES = global <4 x i32> <i32 0, i32 0, i32 0, i32 0>, align 16
51 define void @llvm_mips_bclr_w_test() nounwind {
53 %0 = load <4 x i32>* @llvm_mips_bclr_w_ARG1
54 %1 = load <4 x i32>* @llvm_mips_bclr_w_ARG2
55 %2 = tail call <4 x i32> @llvm.mips.bclr.w(<4 x i32> %0, <4 x i32> %1)
56 store <4 x i32> %2, <4 x i32>* @llvm_mips_bclr_w_RES
60 declare <4 x i32> @llvm.mips.bclr.w(<4 x i32>, <4 x i32>) nounwind
62 ; CHECK: llvm_mips_bclr_w_test:
67 ; CHECK: .size llvm_mips_bclr_w_test
69 @llvm_mips_bclr_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
70 @llvm_mips_bclr_d_ARG2 = global <2 x i64> <i64 2, i64 3>, align 16
71 @llvm_mips_bclr_d_RES = global <2 x i64> <i64 0, i64 0>, align 16
73 define void @llvm_mips_bclr_d_test() nounwind {
75 %0 = load <2 x i64>* @llvm_mips_bclr_d_ARG1
76 %1 = load <2 x i64>* @llvm_mips_bclr_d_ARG2
77 %2 = tail call <2 x i64> @llvm.mips.bclr.d(<2 x i64> %0, <2 x i64> %1)
78 store <2 x i64> %2, <2 x i64>* @llvm_mips_bclr_d_RES
82 declare <2 x i64> @llvm.mips.bclr.d(<2 x i64>, <2 x i64>) nounwind
84 ; CHECK: llvm_mips_bclr_d_test:
89 ; CHECK: .size llvm_mips_bclr_d_test
91 @llvm_mips_binsl_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
92 @llvm_mips_binsl_b_ARG2 = global <16 x i8> <i8 16, i8 17, i8 18, i8 19, i8 20, i8 21, i8 22, i8 23, i8 24, i8 25, i8 26, i8 27, i8 28, i8 29, i8 30, i8 31>, align 16
93 @llvm_mips_binsl_b_RES = global <16 x i8> <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, align 16
95 define void @llvm_mips_binsl_b_test() nounwind {
97 %0 = load <16 x i8>* @llvm_mips_binsl_b_ARG1
98 %1 = load <16 x i8>* @llvm_mips_binsl_b_ARG2
99 %2 = tail call <16 x i8> @llvm.mips.binsl.b(<16 x i8> %0, <16 x i8> %1)
100 store <16 x i8> %2, <16 x i8>* @llvm_mips_binsl_b_RES
104 declare <16 x i8> @llvm.mips.binsl.b(<16 x i8>, <16 x i8>) nounwind
106 ; CHECK: llvm_mips_binsl_b_test:
111 ; CHECK: .size llvm_mips_binsl_b_test
113 @llvm_mips_binsl_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
114 @llvm_mips_binsl_h_ARG2 = global <8 x i16> <i16 8, i16 9, i16 10, i16 11, i16 12, i16 13, i16 14, i16 15>, align 16
115 @llvm_mips_binsl_h_RES = global <8 x i16> <i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>, align 16
117 define void @llvm_mips_binsl_h_test() nounwind {
119 %0 = load <8 x i16>* @llvm_mips_binsl_h_ARG1
120 %1 = load <8 x i16>* @llvm_mips_binsl_h_ARG2
121 %2 = tail call <8 x i16> @llvm.mips.binsl.h(<8 x i16> %0, <8 x i16> %1)
122 store <8 x i16> %2, <8 x i16>* @llvm_mips_binsl_h_RES
126 declare <8 x i16> @llvm.mips.binsl.h(<8 x i16>, <8 x i16>) nounwind
128 ; CHECK: llvm_mips_binsl_h_test:
133 ; CHECK: .size llvm_mips_binsl_h_test
135 @llvm_mips_binsl_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
136 @llvm_mips_binsl_w_ARG2 = global <4 x i32> <i32 4, i32 5, i32 6, i32 7>, align 16
137 @llvm_mips_binsl_w_RES = global <4 x i32> <i32 0, i32 0, i32 0, i32 0>, align 16
139 define void @llvm_mips_binsl_w_test() nounwind {
141 %0 = load <4 x i32>* @llvm_mips_binsl_w_ARG1
142 %1 = load <4 x i32>* @llvm_mips_binsl_w_ARG2
143 %2 = tail call <4 x i32> @llvm.mips.binsl.w(<4 x i32> %0, <4 x i32> %1)
144 store <4 x i32> %2, <4 x i32>* @llvm_mips_binsl_w_RES
148 declare <4 x i32> @llvm.mips.binsl.w(<4 x i32>, <4 x i32>) nounwind
150 ; CHECK: llvm_mips_binsl_w_test:
155 ; CHECK: .size llvm_mips_binsl_w_test
157 @llvm_mips_binsl_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
158 @llvm_mips_binsl_d_ARG2 = global <2 x i64> <i64 2, i64 3>, align 16
159 @llvm_mips_binsl_d_RES = global <2 x i64> <i64 0, i64 0>, align 16
161 define void @llvm_mips_binsl_d_test() nounwind {
163 %0 = load <2 x i64>* @llvm_mips_binsl_d_ARG1
164 %1 = load <2 x i64>* @llvm_mips_binsl_d_ARG2
165 %2 = tail call <2 x i64> @llvm.mips.binsl.d(<2 x i64> %0, <2 x i64> %1)
166 store <2 x i64> %2, <2 x i64>* @llvm_mips_binsl_d_RES
170 declare <2 x i64> @llvm.mips.binsl.d(<2 x i64>, <2 x i64>) nounwind
172 ; CHECK: llvm_mips_binsl_d_test:
177 ; CHECK: .size llvm_mips_binsl_d_test
179 @llvm_mips_binsr_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
180 @llvm_mips_binsr_b_ARG2 = global <16 x i8> <i8 16, i8 17, i8 18, i8 19, i8 20, i8 21, i8 22, i8 23, i8 24, i8 25, i8 26, i8 27, i8 28, i8 29, i8 30, i8 31>, align 16
181 @llvm_mips_binsr_b_RES = global <16 x i8> <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, align 16
183 define void @llvm_mips_binsr_b_test() nounwind {
185 %0 = load <16 x i8>* @llvm_mips_binsr_b_ARG1
186 %1 = load <16 x i8>* @llvm_mips_binsr_b_ARG2
187 %2 = tail call <16 x i8> @llvm.mips.binsr.b(<16 x i8> %0, <16 x i8> %1)
188 store <16 x i8> %2, <16 x i8>* @llvm_mips_binsr_b_RES
192 declare <16 x i8> @llvm.mips.binsr.b(<16 x i8>, <16 x i8>) nounwind
194 ; CHECK: llvm_mips_binsr_b_test:
199 ; CHECK: .size llvm_mips_binsr_b_test
201 @llvm_mips_binsr_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
202 @llvm_mips_binsr_h_ARG2 = global <8 x i16> <i16 8, i16 9, i16 10, i16 11, i16 12, i16 13, i16 14, i16 15>, align 16
203 @llvm_mips_binsr_h_RES = global <8 x i16> <i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>, align 16
205 define void @llvm_mips_binsr_h_test() nounwind {
207 %0 = load <8 x i16>* @llvm_mips_binsr_h_ARG1
208 %1 = load <8 x i16>* @llvm_mips_binsr_h_ARG2
209 %2 = tail call <8 x i16> @llvm.mips.binsr.h(<8 x i16> %0, <8 x i16> %1)
210 store <8 x i16> %2, <8 x i16>* @llvm_mips_binsr_h_RES
214 declare <8 x i16> @llvm.mips.binsr.h(<8 x i16>, <8 x i16>) nounwind
216 ; CHECK: llvm_mips_binsr_h_test:
221 ; CHECK: .size llvm_mips_binsr_h_test
223 @llvm_mips_binsr_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
224 @llvm_mips_binsr_w_ARG2 = global <4 x i32> <i32 4, i32 5, i32 6, i32 7>, align 16
225 @llvm_mips_binsr_w_RES = global <4 x i32> <i32 0, i32 0, i32 0, i32 0>, align 16
227 define void @llvm_mips_binsr_w_test() nounwind {
229 %0 = load <4 x i32>* @llvm_mips_binsr_w_ARG1
230 %1 = load <4 x i32>* @llvm_mips_binsr_w_ARG2
231 %2 = tail call <4 x i32> @llvm.mips.binsr.w(<4 x i32> %0, <4 x i32> %1)
232 store <4 x i32> %2, <4 x i32>* @llvm_mips_binsr_w_RES
236 declare <4 x i32> @llvm.mips.binsr.w(<4 x i32>, <4 x i32>) nounwind
238 ; CHECK: llvm_mips_binsr_w_test:
243 ; CHECK: .size llvm_mips_binsr_w_test
245 @llvm_mips_binsr_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
246 @llvm_mips_binsr_d_ARG2 = global <2 x i64> <i64 2, i64 3>, align 16
247 @llvm_mips_binsr_d_RES = global <2 x i64> <i64 0, i64 0>, align 16
249 define void @llvm_mips_binsr_d_test() nounwind {
251 %0 = load <2 x i64>* @llvm_mips_binsr_d_ARG1
252 %1 = load <2 x i64>* @llvm_mips_binsr_d_ARG2
253 %2 = tail call <2 x i64> @llvm.mips.binsr.d(<2 x i64> %0, <2 x i64> %1)
254 store <2 x i64> %2, <2 x i64>* @llvm_mips_binsr_d_RES
258 declare <2 x i64> @llvm.mips.binsr.d(<2 x i64>, <2 x i64>) nounwind
260 ; CHECK: llvm_mips_binsr_d_test:
265 ; CHECK: .size llvm_mips_binsr_d_test
267 @llvm_mips_bneg_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
268 @llvm_mips_bneg_b_ARG2 = global <16 x i8> <i8 16, i8 17, i8 18, i8 19, i8 20, i8 21, i8 22, i8 23, i8 24, i8 25, i8 26, i8 27, i8 28, i8 29, i8 30, i8 31>, align 16
269 @llvm_mips_bneg_b_RES = global <16 x i8> <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, align 16
271 define void @llvm_mips_bneg_b_test() nounwind {
273 %0 = load <16 x i8>* @llvm_mips_bneg_b_ARG1
274 %1 = load <16 x i8>* @llvm_mips_bneg_b_ARG2
275 %2 = tail call <16 x i8> @llvm.mips.bneg.b(<16 x i8> %0, <16 x i8> %1)
276 store <16 x i8> %2, <16 x i8>* @llvm_mips_bneg_b_RES
280 declare <16 x i8> @llvm.mips.bneg.b(<16 x i8>, <16 x i8>) nounwind
282 ; CHECK: llvm_mips_bneg_b_test:
287 ; CHECK: .size llvm_mips_bneg_b_test
289 @llvm_mips_bneg_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
290 @llvm_mips_bneg_h_ARG2 = global <8 x i16> <i16 8, i16 9, i16 10, i16 11, i16 12, i16 13, i16 14, i16 15>, align 16
291 @llvm_mips_bneg_h_RES = global <8 x i16> <i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>, align 16
293 define void @llvm_mips_bneg_h_test() nounwind {
295 %0 = load <8 x i16>* @llvm_mips_bneg_h_ARG1
296 %1 = load <8 x i16>* @llvm_mips_bneg_h_ARG2
297 %2 = tail call <8 x i16> @llvm.mips.bneg.h(<8 x i16> %0, <8 x i16> %1)
298 store <8 x i16> %2, <8 x i16>* @llvm_mips_bneg_h_RES
302 declare <8 x i16> @llvm.mips.bneg.h(<8 x i16>, <8 x i16>) nounwind
304 ; CHECK: llvm_mips_bneg_h_test:
309 ; CHECK: .size llvm_mips_bneg_h_test
311 @llvm_mips_bneg_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
312 @llvm_mips_bneg_w_ARG2 = global <4 x i32> <i32 4, i32 5, i32 6, i32 7>, align 16
313 @llvm_mips_bneg_w_RES = global <4 x i32> <i32 0, i32 0, i32 0, i32 0>, align 16
315 define void @llvm_mips_bneg_w_test() nounwind {
317 %0 = load <4 x i32>* @llvm_mips_bneg_w_ARG1
318 %1 = load <4 x i32>* @llvm_mips_bneg_w_ARG2
319 %2 = tail call <4 x i32> @llvm.mips.bneg.w(<4 x i32> %0, <4 x i32> %1)
320 store <4 x i32> %2, <4 x i32>* @llvm_mips_bneg_w_RES
324 declare <4 x i32> @llvm.mips.bneg.w(<4 x i32>, <4 x i32>) nounwind
326 ; CHECK: llvm_mips_bneg_w_test:
331 ; CHECK: .size llvm_mips_bneg_w_test
333 @llvm_mips_bneg_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
334 @llvm_mips_bneg_d_ARG2 = global <2 x i64> <i64 2, i64 3>, align 16
335 @llvm_mips_bneg_d_RES = global <2 x i64> <i64 0, i64 0>, align 16
337 define void @llvm_mips_bneg_d_test() nounwind {
339 %0 = load <2 x i64>* @llvm_mips_bneg_d_ARG1
340 %1 = load <2 x i64>* @llvm_mips_bneg_d_ARG2
341 %2 = tail call <2 x i64> @llvm.mips.bneg.d(<2 x i64> %0, <2 x i64> %1)
342 store <2 x i64> %2, <2 x i64>* @llvm_mips_bneg_d_RES
346 declare <2 x i64> @llvm.mips.bneg.d(<2 x i64>, <2 x i64>) nounwind
348 ; CHECK: llvm_mips_bneg_d_test:
353 ; CHECK: .size llvm_mips_bneg_d_test
355 @llvm_mips_bset_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
356 @llvm_mips_bset_b_ARG2 = global <16 x i8> <i8 16, i8 17, i8 18, i8 19, i8 20, i8 21, i8 22, i8 23, i8 24, i8 25, i8 26, i8 27, i8 28, i8 29, i8 30, i8 31>, align 16
357 @llvm_mips_bset_b_RES = global <16 x i8> <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, align 16
359 define void @llvm_mips_bset_b_test() nounwind {
361 %0 = load <16 x i8>* @llvm_mips_bset_b_ARG1
362 %1 = load <16 x i8>* @llvm_mips_bset_b_ARG2
363 %2 = tail call <16 x i8> @llvm.mips.bset.b(<16 x i8> %0, <16 x i8> %1)
364 store <16 x i8> %2, <16 x i8>* @llvm_mips_bset_b_RES
368 declare <16 x i8> @llvm.mips.bset.b(<16 x i8>, <16 x i8>) nounwind
370 ; CHECK: llvm_mips_bset_b_test:
375 ; CHECK: .size llvm_mips_bset_b_test
377 @llvm_mips_bset_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
378 @llvm_mips_bset_h_ARG2 = global <8 x i16> <i16 8, i16 9, i16 10, i16 11, i16 12, i16 13, i16 14, i16 15>, align 16
379 @llvm_mips_bset_h_RES = global <8 x i16> <i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>, align 16
381 define void @llvm_mips_bset_h_test() nounwind {
383 %0 = load <8 x i16>* @llvm_mips_bset_h_ARG1
384 %1 = load <8 x i16>* @llvm_mips_bset_h_ARG2
385 %2 = tail call <8 x i16> @llvm.mips.bset.h(<8 x i16> %0, <8 x i16> %1)
386 store <8 x i16> %2, <8 x i16>* @llvm_mips_bset_h_RES
390 declare <8 x i16> @llvm.mips.bset.h(<8 x i16>, <8 x i16>) nounwind
392 ; CHECK: llvm_mips_bset_h_test:
397 ; CHECK: .size llvm_mips_bset_h_test
399 @llvm_mips_bset_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
400 @llvm_mips_bset_w_ARG2 = global <4 x i32> <i32 4, i32 5, i32 6, i32 7>, align 16
401 @llvm_mips_bset_w_RES = global <4 x i32> <i32 0, i32 0, i32 0, i32 0>, align 16
403 define void @llvm_mips_bset_w_test() nounwind {
405 %0 = load <4 x i32>* @llvm_mips_bset_w_ARG1
406 %1 = load <4 x i32>* @llvm_mips_bset_w_ARG2
407 %2 = tail call <4 x i32> @llvm.mips.bset.w(<4 x i32> %0, <4 x i32> %1)
408 store <4 x i32> %2, <4 x i32>* @llvm_mips_bset_w_RES
412 declare <4 x i32> @llvm.mips.bset.w(<4 x i32>, <4 x i32>) nounwind
414 ; CHECK: llvm_mips_bset_w_test:
419 ; CHECK: .size llvm_mips_bset_w_test
421 @llvm_mips_bset_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
422 @llvm_mips_bset_d_ARG2 = global <2 x i64> <i64 2, i64 3>, align 16
423 @llvm_mips_bset_d_RES = global <2 x i64> <i64 0, i64 0>, align 16
425 define void @llvm_mips_bset_d_test() nounwind {
427 %0 = load <2 x i64>* @llvm_mips_bset_d_ARG1
428 %1 = load <2 x i64>* @llvm_mips_bset_d_ARG2
429 %2 = tail call <2 x i64> @llvm.mips.bset.d(<2 x i64> %0, <2 x i64> %1)
430 store <2 x i64> %2, <2 x i64>* @llvm_mips_bset_d_RES
434 declare <2 x i64> @llvm.mips.bset.d(<2 x i64>, <2 x i64>) nounwind
436 ; CHECK: llvm_mips_bset_d_test:
441 ; CHECK: .size llvm_mips_bset_d_test