1 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
2 ; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+xtheadvdot \
3 ; RUN: -verify-machineinstrs | FileCheck %s
4 ; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+xtheadvdot \
5 ; RUN: -verify-machineinstrs | FileCheck %s
7 declare <vscale x 1 x i32> @llvm.riscv.th.vmaqa.nxv1i32.nxv4i8(
14 define <vscale x 1 x i32> @intrinsic_th_vmaqa_vv_nxv1i32_nxv4i8_nxv4i8(<vscale x 1 x i32> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, iXLen %3) nounwind {
15 ; CHECK-LABEL: intrinsic_th_vmaqa_vv_nxv1i32_nxv4i8_nxv4i8:
16 ; CHECK: # %bb.0: # %entry
17 ; CHECK-NEXT: vsetvli zero, a0, e32, mf2, tu, ma
18 ; CHECK-NEXT: th.vmaqa.vv v8, v9, v10
21 %a = call <vscale x 1 x i32> @llvm.riscv.th.vmaqa.nxv1i32.nxv4i8(
22 <vscale x 1 x i32> %0,
27 ret <vscale x 1 x i32> %a
30 declare <vscale x 1 x i32> @llvm.riscv.th.vmaqa.mask.nxv1i32.nxv4i8(
37 define <vscale x 1 x i32> @intrinsic_th_vmaqa_mask_vv_nxv1i32_nxv4i8_nxv4i8(<vscale x 1 x i32> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
38 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vv_nxv1i32_nxv4i8_nxv4i8:
39 ; CHECK: # %bb.0: # %entry
40 ; CHECK-NEXT: vsetvli zero, a0, e32, mf2, tu, mu
41 ; CHECK-NEXT: th.vmaqa.vv v8, v9, v10, v0.t
44 %a = call <vscale x 1 x i32> @llvm.riscv.th.vmaqa.mask.nxv1i32.nxv4i8(
45 <vscale x 1 x i32> %0,
51 ret <vscale x 1 x i32> %a
54 declare <vscale x 2 x i32> @llvm.riscv.th.vmaqa.nxv2i32.nxv8i8(
61 define <vscale x 2 x i32> @intrinsic_th_vmaqa_vv_nxv2i32_nxv8i8_nxv8i8(<vscale x 2 x i32> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, iXLen %3) nounwind {
62 ; CHECK-LABEL: intrinsic_th_vmaqa_vv_nxv2i32_nxv8i8_nxv8i8:
63 ; CHECK: # %bb.0: # %entry
64 ; CHECK-NEXT: vsetvli zero, a0, e32, m1, tu, ma
65 ; CHECK-NEXT: th.vmaqa.vv v8, v9, v10
68 %a = call <vscale x 2 x i32> @llvm.riscv.th.vmaqa.nxv2i32.nxv8i8(
69 <vscale x 2 x i32> %0,
74 ret <vscale x 2 x i32> %a
77 declare <vscale x 2 x i32> @llvm.riscv.th.vmaqa.mask.nxv2i32.nxv8i8(
84 define <vscale x 2 x i32> @intrinsic_th_vmaqa_mask_vv_nxv2i32_nxv8i8_nxv8i8(<vscale x 2 x i32> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
85 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vv_nxv2i32_nxv8i8_nxv8i8:
86 ; CHECK: # %bb.0: # %entry
87 ; CHECK-NEXT: vsetvli zero, a0, e32, m1, tu, mu
88 ; CHECK-NEXT: th.vmaqa.vv v8, v9, v10, v0.t
91 %a = call <vscale x 2 x i32> @llvm.riscv.th.vmaqa.mask.nxv2i32.nxv8i8(
92 <vscale x 2 x i32> %0,
98 ret <vscale x 2 x i32> %a
101 declare <vscale x 4 x i32> @llvm.riscv.th.vmaqa.nxv4i32.nxv16i8(
108 define <vscale x 4 x i32> @intrinsic_th_vmaqa_vv_nxv4i32_nxv16i8_nxv16i8(<vscale x 4 x i32> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, iXLen %3) nounwind {
109 ; CHECK-LABEL: intrinsic_th_vmaqa_vv_nxv4i32_nxv16i8_nxv16i8:
110 ; CHECK: # %bb.0: # %entry
111 ; CHECK-NEXT: vsetvli zero, a0, e32, m2, tu, ma
112 ; CHECK-NEXT: th.vmaqa.vv v8, v10, v12
115 %a = call <vscale x 4 x i32> @llvm.riscv.th.vmaqa.nxv4i32.nxv16i8(
116 <vscale x 4 x i32> %0,
117 <vscale x 16 x i8> %1,
118 <vscale x 16 x i8> %2,
121 ret <vscale x 4 x i32> %a
124 declare <vscale x 4 x i32> @llvm.riscv.th.vmaqa.mask.nxv4i32.nxv16i8(
131 define <vscale x 4 x i32> @intrinsic_th_vmaqa_mask_vv_nxv4i32_nxv16i8_nxv16i8(<vscale x 4 x i32> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
132 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vv_nxv4i32_nxv16i8_nxv16i8:
133 ; CHECK: # %bb.0: # %entry
134 ; CHECK-NEXT: vsetvli zero, a0, e32, m2, tu, mu
135 ; CHECK-NEXT: th.vmaqa.vv v8, v10, v12, v0.t
138 %a = call <vscale x 4 x i32> @llvm.riscv.th.vmaqa.mask.nxv4i32.nxv16i8(
139 <vscale x 4 x i32> %0,
140 <vscale x 16 x i8> %1,
141 <vscale x 16 x i8> %2,
142 <vscale x 16 x i1> %3,
145 ret <vscale x 4 x i32> %a
148 declare <vscale x 8 x i32> @llvm.riscv.th.vmaqa.nxv8i32.nxv32i8(
155 define <vscale x 8 x i32> @intrinsic_th_vmaqa_vv_nxv8i32_nxv32i8_nxv32i8(<vscale x 8 x i32> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, iXLen %3) nounwind {
156 ; CHECK-LABEL: intrinsic_th_vmaqa_vv_nxv8i32_nxv32i8_nxv32i8:
157 ; CHECK: # %bb.0: # %entry
158 ; CHECK-NEXT: vsetvli zero, a0, e32, m4, tu, ma
159 ; CHECK-NEXT: th.vmaqa.vv v8, v12, v16
162 %a = call <vscale x 8 x i32> @llvm.riscv.th.vmaqa.nxv8i32.nxv32i8(
163 <vscale x 8 x i32> %0,
164 <vscale x 32 x i8> %1,
165 <vscale x 32 x i8> %2,
168 ret <vscale x 8 x i32> %a
171 declare <vscale x 8 x i32> @llvm.riscv.th.vmaqa.mask.nxv8i32.nxv32i8(
178 define <vscale x 8 x i32> @intrinsic_th_vmaqa_mask_vv_nxv8i32_nxv32i8_nxv32i8(<vscale x 8 x i32> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, iXLen %4) nounwind {
179 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vv_nxv8i32_nxv32i8_nxv32i8:
180 ; CHECK: # %bb.0: # %entry
181 ; CHECK-NEXT: vsetvli zero, a0, e32, m4, tu, mu
182 ; CHECK-NEXT: th.vmaqa.vv v8, v12, v16, v0.t
185 %a = call <vscale x 8 x i32> @llvm.riscv.th.vmaqa.mask.nxv8i32.nxv32i8(
186 <vscale x 8 x i32> %0,
187 <vscale x 32 x i8> %1,
188 <vscale x 32 x i8> %2,
189 <vscale x 32 x i1> %3,
192 ret <vscale x 8 x i32> %a
196 declare <vscale x 1 x i32> @llvm.riscv.th.vmaqa.nxv1i32.i8(
203 define <vscale x 1 x i32> @intrinsic_th_vmaqa_vx_nxv1i32_i8_nxv4i8(<vscale x 1 x i32> %0, i8 %1, <vscale x 4 x i8> %2, iXLen %3) nounwind {
204 ; CHECK-LABEL: intrinsic_th_vmaqa_vx_nxv1i32_i8_nxv4i8:
205 ; CHECK: # %bb.0: # %entry
206 ; CHECK-NEXT: vsetvli zero, a1, e32, mf2, tu, ma
207 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v9
210 %a = call <vscale x 1 x i32> @llvm.riscv.th.vmaqa.nxv1i32.i8(
211 <vscale x 1 x i32> %0,
213 <vscale x 4 x i8> %2,
216 ret <vscale x 1 x i32> %a
219 declare <vscale x 1 x i32> @llvm.riscv.th.vmaqa.mask.nxv1i32.i8(
226 define <vscale x 1 x i32> @intrinsic_th_vmaqa_mask_vx_nxv1i32_i8_nxv4i8(<vscale x 1 x i32> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
227 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vx_nxv1i32_i8_nxv4i8:
228 ; CHECK: # %bb.0: # %entry
229 ; CHECK-NEXT: vsetvli zero, a1, e32, mf2, tu, mu
230 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v9, v0.t
233 %a = call <vscale x 1 x i32> @llvm.riscv.th.vmaqa.mask.nxv1i32.i8(
234 <vscale x 1 x i32> %0,
236 <vscale x 4 x i8> %2,
237 <vscale x 4 x i1> %3,
240 ret <vscale x 1 x i32> %a
243 declare <vscale x 2 x i32> @llvm.riscv.th.vmaqa.nxv2i32.i8(
250 define <vscale x 2 x i32> @intrinsic_th_vmaqa_vx_nxv2i32_i8_nxv8i8(<vscale x 2 x i32> %0, i8 %1, <vscale x 8 x i8> %2, iXLen %3) nounwind {
251 ; CHECK-LABEL: intrinsic_th_vmaqa_vx_nxv2i32_i8_nxv8i8:
252 ; CHECK: # %bb.0: # %entry
253 ; CHECK-NEXT: vsetvli zero, a1, e32, m1, tu, ma
254 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v9
257 %a = call <vscale x 2 x i32> @llvm.riscv.th.vmaqa.nxv2i32.i8(
258 <vscale x 2 x i32> %0,
260 <vscale x 8 x i8> %2,
263 ret <vscale x 2 x i32> %a
266 declare <vscale x 2 x i32> @llvm.riscv.th.vmaqa.mask.nxv2i32.i8(
273 define <vscale x 2 x i32> @intrinsic_th_vmaqa_mask_vx_nxv2i32_i8_nxv8i8(<vscale x 2 x i32> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
274 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vx_nxv2i32_i8_nxv8i8:
275 ; CHECK: # %bb.0: # %entry
276 ; CHECK-NEXT: vsetvli zero, a1, e32, m1, tu, mu
277 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v9, v0.t
280 %a = call <vscale x 2 x i32> @llvm.riscv.th.vmaqa.mask.nxv2i32.i8(
281 <vscale x 2 x i32> %0,
283 <vscale x 8 x i8> %2,
284 <vscale x 8 x i1> %3,
287 ret <vscale x 2 x i32> %a
290 declare <vscale x 4 x i32> @llvm.riscv.th.vmaqa.nxv4i32.i8(
297 define <vscale x 4 x i32> @intrinsic_th_vmaqa_vx_nxv4i32_i8_nxv16i8(<vscale x 4 x i32> %0, i8 %1, <vscale x 16 x i8> %2, iXLen %3) nounwind {
298 ; CHECK-LABEL: intrinsic_th_vmaqa_vx_nxv4i32_i8_nxv16i8:
299 ; CHECK: # %bb.0: # %entry
300 ; CHECK-NEXT: vsetvli zero, a1, e32, m2, tu, ma
301 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v10
304 %a = call <vscale x 4 x i32> @llvm.riscv.th.vmaqa.nxv4i32.i8(
305 <vscale x 4 x i32> %0,
307 <vscale x 16 x i8> %2,
310 ret <vscale x 4 x i32> %a
313 declare <vscale x 4 x i32> @llvm.riscv.th.vmaqa.mask.nxv4i32.i8(
320 define <vscale x 4 x i32> @intrinsic_th_vmaqa_mask_vx_nxv4i32_i8_nxv16i8(<vscale x 4 x i32> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
321 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vx_nxv4i32_i8_nxv16i8:
322 ; CHECK: # %bb.0: # %entry
323 ; CHECK-NEXT: vsetvli zero, a1, e32, m2, tu, mu
324 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v10, v0.t
327 %a = call <vscale x 4 x i32> @llvm.riscv.th.vmaqa.mask.nxv4i32.i8(
328 <vscale x 4 x i32> %0,
330 <vscale x 16 x i8> %2,
331 <vscale x 16 x i1> %3,
334 ret <vscale x 4 x i32> %a
337 declare <vscale x 8 x i32> @llvm.riscv.th.vmaqa.nxv8i32.i8(
344 define <vscale x 8 x i32> @intrinsic_th_vmaqa_vx_nxv8i32_i8_nxv32i8(<vscale x 8 x i32> %0, i8 %1, <vscale x 32 x i8> %2, iXLen %3) nounwind {
345 ; CHECK-LABEL: intrinsic_th_vmaqa_vx_nxv8i32_i8_nxv32i8:
346 ; CHECK: # %bb.0: # %entry
347 ; CHECK-NEXT: vsetvli zero, a1, e32, m4, tu, ma
348 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v12
351 %a = call <vscale x 8 x i32> @llvm.riscv.th.vmaqa.nxv8i32.i8(
352 <vscale x 8 x i32> %0,
354 <vscale x 32 x i8> %2,
357 ret <vscale x 8 x i32> %a
360 declare <vscale x 8 x i32> @llvm.riscv.th.vmaqa.mask.nxv8i32.i8(
367 define <vscale x 8 x i32> @intrinsic_th_vmaqa_mask_vx_nxv8i32_i8_nxv32i8(<vscale x 8 x i32> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, iXLen %4) nounwind {
368 ; CHECK-LABEL: intrinsic_th_vmaqa_mask_vx_nxv8i32_i8_nxv32i8:
369 ; CHECK: # %bb.0: # %entry
370 ; CHECK-NEXT: vsetvli zero, a1, e32, m4, tu, mu
371 ; CHECK-NEXT: th.vmaqa.vx v8, a0, v12, v0.t
374 %a = call <vscale x 8 x i32> @llvm.riscv.th.vmaqa.mask.nxv8i32.i8(
375 <vscale x 8 x i32> %0,
377 <vscale x 32 x i8> %2,
378 <vscale x 32 x i1> %3,
381 ret <vscale x 8 x i32> %a