clang/test/CodeGen/AArch64/sve-intrinsics/acle_sve_matmul_fp32.c

   1 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
   2 // RUN: %clang_cc1 -fclang-abi-compat=latest -target-feature +f32mm -triple aarch64 -target-feature +sve -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s
   3 // RUN: %clang_cc1 -fclang-abi-compat=latest -target-feature +f32mm -triple aarch64 -target-feature +sve -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 -fclang-abi-compat=latest -target-feature +f32mm -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sve -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s
   5 // RUN: %clang_cc1 -fclang-abi-compat=latest -target-feature +f32mm -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sve -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
   6
   7 // REQUIRES: aarch64-registered-target
   8
   9 #include <arm_sve.h>
  10
  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
  17
  18 // CHECK-LABEL: @test_svmmla_f32(
  19 // CHECK-NEXT:  entry:
  20 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.fmmla.nxv4f32(<vscale x 4 x float> [[X:%.*]], <vscale x 4 x float> [[Y:%.*]], <vscale x 4 x float> [[Z:%.*]])
  21 // CHECK-NEXT:    ret <vscale x 4 x float> [[TMP0]]
  22 //
  23 // CPP-CHECK-LABEL: @_Z15test_svmmla_f32u13__SVFloat32_tS_S_(
  24 // CPP-CHECK-NEXT:  entry:
  25 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.fmmla.nxv4f32(<vscale x 4 x float> [[X:%.*]], <vscale x 4 x float> [[Y:%.*]], <vscale x 4 x float> [[Z:%.*]])
  26 // CPP-CHECK-NEXT:    ret <vscale x 4 x float> [[TMP0]]
  27 //
  28 svfloat32_t test_svmmla_f32(svfloat32_t x, svfloat32_t y, svfloat32_t z) {
  29   return SVE_ACLE_FUNC(svmmla,_f32,,)(x, y, z);
  30 }