Run DCE after a LoopFlatten test to reduce spurious output [nfc]
[llvm-project.git] / clang / test / CodeGen / aarch64-sve-intrinsics / acle_sve_bfmmla.c
blobb88674d6668a5de9624d6f07a31cd37f9ee046fc
1 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
2 // RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s
3 // RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
4 // RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s
5 // RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
7 // REQUIRES: aarch64-registered-target
9 #include <arm_sve.h>
11 #ifdef SVE_OVERLOADED_FORMS
12 // A simple used,unused... macro, long enough to represent any SVE builtin.
13 #define SVE_ACLE_FUNC(A1, A2_UNUSED, A3, A4_UNUSED) A1##A3
14 #else
15 #define SVE_ACLE_FUNC(A1, A2, A3, A4) A1##A2##A3##A4
16 #endif
18 // CHECK-LABEL: @test_bfmmla_f32(
19 // CHECK-NEXT: entry:
20 // CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.bfmmla(<vscale x 4 x float> [[X:%.*]], <vscale x 8 x bfloat> [[Y:%.*]], <vscale x 8 x bfloat> [[Z:%.*]])
21 // CHECK-NEXT: ret <vscale x 4 x float> [[TMP0]]
23 // CPP-CHECK-LABEL: @_Z15test_bfmmla_f32u13__SVFloat32_tu14__SVBfloat16_tS0_(
24 // CPP-CHECK-NEXT: entry:
25 // CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.bfmmla(<vscale x 4 x float> [[X:%.*]], <vscale x 8 x bfloat> [[Y:%.*]], <vscale x 8 x bfloat> [[Z:%.*]])
26 // CPP-CHECK-NEXT: ret <vscale x 4 x float> [[TMP0]]
28 svfloat32_t test_bfmmla_f32(svfloat32_t x, svbfloat16_t y, svbfloat16_t z) {
29 return SVE_ACLE_FUNC(svbfmmla, _f32, , )(x, y, z);