forked from OSchip/llvm-project
31 lines
2.0 KiB
C
31 lines
2.0 KiB
C
// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
|
|
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
|
|
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
|
|
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
|
|
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
|
|
|
|
// REQUIRES: aarch64-registered-target
|
|
|
|
#include <arm_sve.h>
|
|
|
|
#ifdef SVE_OVERLOADED_FORMS
|
|
// A simple used,unused... macro, long enough to represent any SVE builtin.
|
|
#define SVE_ACLE_FUNC(A1, A2_UNUSED, A3, A4_UNUSED) A1##A3
|
|
#else
|
|
#define SVE_ACLE_FUNC(A1, A2, A3, A4) A1##A2##A3##A4
|
|
#endif
|
|
|
|
// CHECK-LABEL: @test_bfmmla_f32(
|
|
// CHECK-NEXT: entry:
|
|
// CHECK-NEXT: [[TMP0:%.*]] = 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:%.*]])
|
|
// CHECK-NEXT: ret <vscale x 4 x float> [[TMP0]]
|
|
//
|
|
// CPP-CHECK-LABEL: @_Z15test_bfmmla_f32u13__SVFloat32_tu14__SVBFloat16_tu14__SVBFloat16_t(
|
|
// CPP-CHECK-NEXT: entry:
|
|
// CPP-CHECK-NEXT: [[TMP0:%.*]] = 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:%.*]])
|
|
// CPP-CHECK-NEXT: ret <vscale x 4 x float> [[TMP0]]
|
|
//
|
|
svfloat32_t test_bfmmla_f32(svfloat32_t x, svbfloat16_t y, svbfloat16_t z) {
|
|
return SVE_ACLE_FUNC(svbfmmla, _f32, , )(x, y, z);
|
|
}
|