clang/test/CodeGen/AArch64/sme-intrinsics/acle_sme_mopa-za64.c

   1 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 3
   2 // REQUIRES: aarch64-registered-target
   3 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-f64f64 -target-feature +sme-i16i64 -target-feature +bf16 -O1 -Werror -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK,CHECK-C
   4 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-f64f64 -target-feature +sme-i16i64 -target-feature +bf16 -O1 -Werror -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefixes=CHECK,CHECK-CXX
   5 // RUN: %clang_cc1 -fclang-abi-compat=latest -DSME_OVERLOADED_FORMS -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-f64f64 -target-feature +sme-i16i64 -target-feature +bf16 -O1 -Werror -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK,CHECK-C
   6 // RUN: %clang_cc1 -fclang-abi-compat=latest -DSME_OVERLOADED_FORMS -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-f64f64 -target-feature +sme-i16i64 -target-feature +bf16 -O1 -Werror -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefixes=CHECK,CHECK-CXX
   7 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-f64f64 -target-feature +sme-i16i64 -target-feature +bf16 -S -O1 -Werror -o /dev/null %s
   8
   9 #include <arm_sme.h>
  10
  11 #ifdef SME_OVERLOADED_FORMS
  12 #define SME_ACLE_FUNC(A1,A2_UNUSED,A3) A1##A3
  13 #else
  14 #define SME_ACLE_FUNC(A1,A2,A3) A1##A2##A3
  15 #endif
  16
  17 // CHECK-C-LABEL: define dso_local void @test_svmopa_za64_s16(
  18 // CHECK-C-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
  19 // CHECK-C-NEXT:  entry:
  20 // CHECK-C-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
  21 // CHECK-C-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
  22 // CHECK-C-NEXT:    tail call void @llvm.aarch64.sme.smopa.wide.nxv8i16(i32 7, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
  23 // CHECK-C-NEXT:    ret void
  24 //
  25 // CHECK-CXX-LABEL: define dso_local void @_Z20test_svmopa_za64_s16u10__SVBool_tS_u11__SVInt16_tS0_(
  26 // CHECK-CXX-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
  27 // CHECK-CXX-NEXT:  entry:
  28 // CHECK-CXX-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
  29 // CHECK-CXX-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
  30 // CHECK-CXX-NEXT:    tail call void @llvm.aarch64.sme.smopa.wide.nxv8i16(i32 7, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
  31 // CHECK-CXX-NEXT:    ret void
  32 //
  33 void test_svmopa_za64_s16(svbool_t pn, svbool_t pm, svint16_t zn, svint16_t zm) __arm_streaming __arm_inout("za") {
  34   SME_ACLE_FUNC(svmopa_za64, _s16, _m)(7, pn, pm, zn, zm);
  35 }
  36
  37 // CHECK-C-LABEL: define dso_local void @test_svmopa_za64_u16(
  38 // CHECK-C-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  39 // CHECK-C-NEXT:  entry:
  40 // CHECK-C-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
  41 // CHECK-C-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
  42 // CHECK-C-NEXT:    tail call void @llvm.aarch64.sme.umopa.wide.nxv8i16(i32 0, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
  43 // CHECK-C-NEXT:    ret void
  44 //
  45 // CHECK-CXX-LABEL: define dso_local void @_Z20test_svmopa_za64_u16u10__SVBool_tS_u12__SVUint16_tS0_(
  46 // CHECK-CXX-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  47 // CHECK-CXX-NEXT:  entry:
  48 // CHECK-CXX-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
  49 // CHECK-CXX-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
  50 // CHECK-CXX-NEXT:    tail call void @llvm.aarch64.sme.umopa.wide.nxv8i16(i32 0, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
  51 // CHECK-CXX-NEXT:    ret void
  52 //
  53 void test_svmopa_za64_u16(svbool_t pn, svbool_t pm, svuint16_t zn, svuint16_t zm) __arm_streaming __arm_inout("za") {
  54   SME_ACLE_FUNC(svmopa_za64, _u16, _m)(0, pn, pm, zn, zm);
  55 }
  56
  57 // CHECK-C-LABEL: define dso_local void @test_svmopa_za64_f64(
  58 // CHECK-C-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 2 x double> [[ZN:%.*]], <vscale x 2 x double> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  59 // CHECK-C-NEXT:  entry:
  60 // CHECK-C-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PN]])
  61 // CHECK-C-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PM]])
  62 // CHECK-C-NEXT:    tail call void @llvm.aarch64.sme.mopa.nxv2f64(i32 7, <vscale x 2 x i1> [[TMP0]], <vscale x 2 x i1> [[TMP1]], <vscale x 2 x double> [[ZN]], <vscale x 2 x double> [[ZM]])
  63 // CHECK-C-NEXT:    ret void
  64 //
  65 // CHECK-CXX-LABEL: define dso_local void @_Z20test_svmopa_za64_f64u10__SVBool_tS_u13__SVFloat64_tS0_(
  66 // CHECK-CXX-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 2 x double> [[ZN:%.*]], <vscale x 2 x double> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  67 // CHECK-CXX-NEXT:  entry:
  68 // CHECK-CXX-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PN]])
  69 // CHECK-CXX-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PM]])
  70 // CHECK-CXX-NEXT:    tail call void @llvm.aarch64.sme.mopa.nxv2f64(i32 7, <vscale x 2 x i1> [[TMP0]], <vscale x 2 x i1> [[TMP1]], <vscale x 2 x double> [[ZN]], <vscale x 2 x double> [[ZM]])
  71 // CHECK-CXX-NEXT:    ret void
  72 //
  73 void test_svmopa_za64_f64(svbool_t pn, svbool_t pm, svfloat64_t zn, svfloat64_t zm) __arm_streaming __arm_inout("za") {
  74   SME_ACLE_FUNC(svmopa_za64, _f64, _m)(7, pn, pm, zn, zm);
  75 }
  76
  77 // CHECK-C-LABEL: define dso_local void @test_svsumopa_za64_s16(
  78 // CHECK-C-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  79 // CHECK-C-NEXT:  entry:
  80 // CHECK-C-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
  81 // CHECK-C-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
  82 // CHECK-C-NEXT:    tail call void @llvm.aarch64.sme.sumopa.wide.nxv8i16(i32 0, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
  83 // CHECK-C-NEXT:    ret void
  84 //
  85 // CHECK-CXX-LABEL: define dso_local void @_Z22test_svsumopa_za64_s16u10__SVBool_tS_u11__SVInt16_tu12__SVUint16_t(
  86 // CHECK-CXX-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  87 // CHECK-CXX-NEXT:  entry:
  88 // CHECK-CXX-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
  89 // CHECK-CXX-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
  90 // CHECK-CXX-NEXT:    tail call void @llvm.aarch64.sme.sumopa.wide.nxv8i16(i32 0, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
  91 // CHECK-CXX-NEXT:    ret void
  92 //
  93 void test_svsumopa_za64_s16(svbool_t pn, svbool_t pm, svint16_t zn, svuint16_t zm) __arm_streaming __arm_inout("za") {
  94  SME_ACLE_FUNC(svsumopa_za64, _s16, _m)(0, pn, pm, zn, zm);
  95 }
  96
  97 // CHECK-C-LABEL: define dso_local void @test_svusmopa_za64_u16(
  98 // CHECK-C-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
  99 // CHECK-C-NEXT:  entry:
 100 // CHECK-C-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
 101 // CHECK-C-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
 102 // CHECK-C-NEXT:    tail call void @llvm.aarch64.sme.usmopa.wide.nxv8i16(i32 7, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
 103 // CHECK-C-NEXT:    ret void
 104 //
 105 // CHECK-CXX-LABEL: define dso_local void @_Z22test_svusmopa_za64_u16u10__SVBool_tS_u12__SVUint16_tu11__SVInt16_t(
 106 // CHECK-CXX-SAME: <vscale x 16 x i1> [[PN:%.*]], <vscale x 16 x i1> [[PM:%.*]], <vscale x 8 x i16> [[ZN:%.*]], <vscale x 8 x i16> [[ZM:%.*]]) local_unnamed_addr #[[ATTR0]] {
 107 // CHECK-CXX-NEXT:  entry:
 108 // CHECK-CXX-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PN]])
 109 // CHECK-CXX-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PM]])
 110 // CHECK-CXX-NEXT:    tail call void @llvm.aarch64.sme.usmopa.wide.nxv8i16(i32 7, <vscale x 8 x i1> [[TMP0]], <vscale x 8 x i1> [[TMP1]], <vscale x 8 x i16> [[ZN]], <vscale x 8 x i16> [[ZM]])
 111 // CHECK-CXX-NEXT:    ret void
 112 //
 113 void test_svusmopa_za64_u16(svbool_t pn, svbool_t pm, svuint16_t zn, svint16_t zm) __arm_streaming __arm_inout("za") {
 114   SME_ACLE_FUNC(svusmopa_za64, _u16, _m)(7, pn, pm, zn, zm);
 115 }
 116 //// NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
 117 // CHECK: {{.*}}