[clang] e51c1d0 - [SveEmitter] Add builtins for svtbl2
Sander de Smalen via cfe-commits
cfe-commits at lists.llvm.org
Wed Jun 17 01:42:37 PDT 2020
Author: Sander de Smalen
Date: 2020-06-17T09:41:38+01:00
New Revision: e51c1d06a9922c3b6ce4b8b2e74126870ade1491
URL: https://github.com/llvm/llvm-project/commit/e51c1d06a9922c3b6ce4b8b2e74126870ade1491
DIFF: https://github.com/llvm/llvm-project/commit/e51c1d06a9922c3b6ce4b8b2e74126870ade1491.diff
LOG: [SveEmitter] Add builtins for svtbl2
Reviewers: david-arm, efriedma, c-rhodes
Reviewed By: c-rhodes
Tags: #clang
Differential Revision: https://reviews.llvm.org/D81462
Added:
clang/test/CodeGen/aarch64-sve2-intrinsics/acle_sve2_tbl2.c
Modified:
clang/include/clang/Basic/arm_sve.td
clang/lib/CodeGen/CGBuiltin.cpp
Removed:
################################################################################
diff --git a/clang/include/clang/Basic/arm_sve.td b/clang/include/clang/Basic/arm_sve.td
index 8c6abb1c3f4f..a7223f770455 100644
--- a/clang/include/clang/Basic/arm_sve.td
+++ b/clang/include/clang/Basic/arm_sve.td
@@ -1838,6 +1838,7 @@ def SVWHILEWR_D : SInst<"svwhilewr[_{1}]", "Pcc", "lUld", MergeNone, "aarch64_sv
////////////////////////////////////////////////////////////////////////////////
// SVE2 - Extended table lookup/permute
let ArchGuard = "defined(__ARM_FEATURE_SVE2)" in {
+def SVTBL2 : SInst<"svtbl2[_{d}]", "d2u", "csilUcUsUiUlhfd", MergeNone>;
def SVTBX : SInst<"svtbx[_{d}]", "dddu", "csilUcUsUiUlhfd", MergeNone, "aarch64_sve_tbx">;
}
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index 3b3ea5e95705..b81b2a449425 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -8265,6 +8265,29 @@ Value *CodeGenFunction::EmitAArch64SVEBuiltinExpr(unsigned BuiltinID,
Function *F = CGM.getIntrinsic(Intrinsic::vscale, Ty);
return Builder.CreateMul(NumEls, Builder.CreateCall(F));
}
+
+ case SVE::BI__builtin_sve_svtbl2_u8:
+ case SVE::BI__builtin_sve_svtbl2_s8:
+ case SVE::BI__builtin_sve_svtbl2_u16:
+ case SVE::BI__builtin_sve_svtbl2_s16:
+ case SVE::BI__builtin_sve_svtbl2_u32:
+ case SVE::BI__builtin_sve_svtbl2_s32:
+ case SVE::BI__builtin_sve_svtbl2_u64:
+ case SVE::BI__builtin_sve_svtbl2_s64:
+ case SVE::BI__builtin_sve_svtbl2_f16:
+ case SVE::BI__builtin_sve_svtbl2_f32:
+ case SVE::BI__builtin_sve_svtbl2_f64: {
+ SVETypeFlags TF(Builtin->TypeModifier);
+ auto VTy = cast<llvm::VectorType>(getSVEType(TF));
+ auto TupleTy = llvm::VectorType::get(VTy->getElementType(),
+ VTy->getElementCount() * 2);
+ Function *FExtr =
+ CGM.getIntrinsic(Intrinsic::aarch64_sve_tuple_get, {VTy, TupleTy});
+ Value *V0 = Builder.CreateCall(FExtr, {Ops[0], Builder.getInt32(0)});
+ Value *V1 = Builder.CreateCall(FExtr, {Ops[0], Builder.getInt32(1)});
+ Function *F = CGM.getIntrinsic(Intrinsic::aarch64_sve_tbl2, VTy);
+ return Builder.CreateCall(F, {V0, V1, Ops[1]});
+ }
}
/// Should not happen
diff --git a/clang/test/CodeGen/aarch64-sve2-intrinsics/acle_sve2_tbl2.c b/clang/test/CodeGen/aarch64-sve2-intrinsics/acle_sve2_tbl2.c
new file mode 100644
index 000000000000..d74b17f72663
--- /dev/null
+++ b/clang/test/CodeGen/aarch64-sve2-intrinsics/acle_sve2_tbl2.c
@@ -0,0 +1,145 @@
+// RUN: %clang_cc1 -D__ARM_FEATURE_SVE -D__ARM_FEATURE_SVE2 -triple aarch64-none-linux-gnu -target-feature +sve2 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -D__ARM_FEATURE_SVE -D__ARM_FEATURE_SVE2 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve2 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -D__ARM_FEATURE_SVE -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -fsyntax-only -verify -verify-ignore-unexpected=error %s
+// RUN: %clang_cc1 -D__ARM_FEATURE_SVE -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -fsyntax-only -verify=overload -verify-ignore-unexpected=error %s
+
+#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
+
+svint8_t test_svtbl2_s8(svint8x2_t data, svuint8_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_s8
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.tuple.get.nxv16i8.nxv32i8(<vscale x 32 x i8> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.tuple.get.nxv16i8.nxv32i8(<vscale x 32 x i8> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.tbl2.nxv16i8(<vscale x 16 x i8> %[[V0]], <vscale x 16 x i8> %[[V1]], <vscale x 16 x i8> %indices)
+ // CHECK-NEXT: ret <vscale x 16 x i8> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_s8'}}
+ return SVE_ACLE_FUNC(svtbl2,_s8,,)(data, indices);
+}
+
+svint16_t test_svtbl2_s16(svint16x2_t data, svuint16_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_s16
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 8 x i16> @llvm.aarch64.sve.tuple.get.nxv8i16.nxv16i16(<vscale x 16 x i16> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 8 x i16> @llvm.aarch64.sve.tuple.get.nxv8i16.nxv16i16(<vscale x 16 x i16> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 8 x i16> @llvm.aarch64.sve.tbl2.nxv8i16(<vscale x 8 x i16> %[[V0]], <vscale x 8 x i16> %[[V1]], <vscale x 8 x i16> %indices)
+ // CHECK-NEXT: ret <vscale x 8 x i16> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_s16'}}
+ return SVE_ACLE_FUNC(svtbl2,_s16,,)(data, indices);
+}
+
+svint32_t test_svtbl2_s32(svint32x2_t data, svuint32_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_s32
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 4 x i32> @llvm.aarch64.sve.tuple.get.nxv4i32.nxv8i32(<vscale x 8 x i32> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 4 x i32> @llvm.aarch64.sve.tuple.get.nxv4i32.nxv8i32(<vscale x 8 x i32> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 4 x i32> @llvm.aarch64.sve.tbl2.nxv4i32(<vscale x 4 x i32> %[[V0]], <vscale x 4 x i32> %[[V1]], <vscale x 4 x i32> %indices)
+ // CHECK-NEXT: ret <vscale x 4 x i32> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_s32'}}
+ return SVE_ACLE_FUNC(svtbl2,_s32,,)(data, indices);
+}
+
+svint64_t test_svtbl2_s64(svint64x2_t data, svuint64_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_s64
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 2 x i64> @llvm.aarch64.sve.tuple.get.nxv2i64.nxv4i64(<vscale x 4 x i64> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 2 x i64> @llvm.aarch64.sve.tuple.get.nxv2i64.nxv4i64(<vscale x 4 x i64> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 2 x i64> @llvm.aarch64.sve.tbl2.nxv2i64(<vscale x 2 x i64> %[[V0]], <vscale x 2 x i64> %[[V1]], <vscale x 2 x i64> %indices)
+ // CHECK-NEXT: ret <vscale x 2 x i64> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_s64'}}
+ return SVE_ACLE_FUNC(svtbl2,_s64,,)(data, indices);
+}
+
+svuint8_t test_svtbl2_u8(svuint8x2_t data, svuint8_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_u8
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.tuple.get.nxv16i8.nxv32i8(<vscale x 32 x i8> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.tuple.get.nxv16i8.nxv32i8(<vscale x 32 x i8> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.tbl2.nxv16i8(<vscale x 16 x i8> %[[V0]], <vscale x 16 x i8> %[[V1]], <vscale x 16 x i8> %indices)
+ // CHECK-NEXT: ret <vscale x 16 x i8> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_u8'}}
+ return SVE_ACLE_FUNC(svtbl2,_u8,,)(data, indices);
+}
+
+svuint16_t test_svtbl2_u16(svuint16x2_t data, svuint16_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_u16
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 8 x i16> @llvm.aarch64.sve.tuple.get.nxv8i16.nxv16i16(<vscale x 16 x i16> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 8 x i16> @llvm.aarch64.sve.tuple.get.nxv8i16.nxv16i16(<vscale x 16 x i16> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 8 x i16> @llvm.aarch64.sve.tbl2.nxv8i16(<vscale x 8 x i16> %[[V0]], <vscale x 8 x i16> %[[V1]], <vscale x 8 x i16> %indices)
+ // CHECK-NEXT: ret <vscale x 8 x i16> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_u16'}}
+ return SVE_ACLE_FUNC(svtbl2,_u16,,)(data, indices);
+}
+
+svuint32_t test_svtbl2_u32(svuint32x2_t data, svuint32_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_u32
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 4 x i32> @llvm.aarch64.sve.tuple.get.nxv4i32.nxv8i32(<vscale x 8 x i32> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 4 x i32> @llvm.aarch64.sve.tuple.get.nxv4i32.nxv8i32(<vscale x 8 x i32> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 4 x i32> @llvm.aarch64.sve.tbl2.nxv4i32(<vscale x 4 x i32> %[[V0]], <vscale x 4 x i32> %[[V1]], <vscale x 4 x i32> %indices)
+ // CHECK-NEXT: ret <vscale x 4 x i32> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_u32'}}
+ return SVE_ACLE_FUNC(svtbl2,_u32,,)(data, indices);
+}
+
+svuint64_t test_svtbl2_u64(svuint64x2_t data, svuint64_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_u64
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 2 x i64> @llvm.aarch64.sve.tuple.get.nxv2i64.nxv4i64(<vscale x 4 x i64> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 2 x i64> @llvm.aarch64.sve.tuple.get.nxv2i64.nxv4i64(<vscale x 4 x i64> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 2 x i64> @llvm.aarch64.sve.tbl2.nxv2i64(<vscale x 2 x i64> %[[V0]], <vscale x 2 x i64> %[[V1]], <vscale x 2 x i64> %indices)
+ // CHECK-NEXT: ret <vscale x 2 x i64> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_u64'}}
+ return SVE_ACLE_FUNC(svtbl2,_u64,,)(data, indices);
+}
+
+svfloat16_t test_svtbl2_f16(svfloat16x2_t data, svuint16_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_f16
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 8 x half> @llvm.aarch64.sve.tuple.get.nxv8f16.nxv16f16(<vscale x 16 x half> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 8 x half> @llvm.aarch64.sve.tuple.get.nxv8f16.nxv16f16(<vscale x 16 x half> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 8 x half> @llvm.aarch64.sve.tbl2.nxv8f16(<vscale x 8 x half> %[[V0]], <vscale x 8 x half> %[[V1]], <vscale x 8 x i16> %indices)
+ // CHECK-NEXT: ret <vscale x 8 x half> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_f16'}}
+ return SVE_ACLE_FUNC(svtbl2,_f16,,)(data, indices);
+}
+
+svfloat32_t test_svtbl2_f32(svfloat32x2_t data, svuint32_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_f32
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 4 x float> @llvm.aarch64.sve.tuple.get.nxv4f32.nxv8f32(<vscale x 8 x float> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 4 x float> @llvm.aarch64.sve.tuple.get.nxv4f32.nxv8f32(<vscale x 8 x float> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 4 x float> @llvm.aarch64.sve.tbl2.nxv4f32(<vscale x 4 x float> %[[V0]], <vscale x 4 x float> %[[V1]], <vscale x 4 x i32> %indices)
+ // CHECK-NEXT: ret <vscale x 4 x float> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_f32'}}
+ return SVE_ACLE_FUNC(svtbl2,_f32,,)(data, indices);
+}
+
+svfloat64_t test_svtbl2_f64(svfloat64x2_t data, svuint64_t indices)
+{
+ // CHECK-LABEL: test_svtbl2_f64
+ // CHECK-DAG: %[[V0:.*]] = call <vscale x 2 x double> @llvm.aarch64.sve.tuple.get.nxv2f64.nxv4f64(<vscale x 4 x double> %data, i32 0)
+ // CHECK-DAG: %[[V1:.*]] = call <vscale x 2 x double> @llvm.aarch64.sve.tuple.get.nxv2f64.nxv4f64(<vscale x 4 x double> %data, i32 1)
+ // CHECK: %[[INTRINSIC:.*]] = call <vscale x 2 x double> @llvm.aarch64.sve.tbl2.nxv2f64(<vscale x 2 x double> %[[V0]], <vscale x 2 x double> %[[V1]], <vscale x 2 x i64> %indices)
+ // CHECK-NEXT: ret <vscale x 2 x double> %[[INTRINSIC]]
+ // overload-warning at +2 {{implicit declaration of function 'svtbl2'}}
+ // expected-warning at +1 {{implicit declaration of function 'svtbl2_f64'}}
+ return SVE_ACLE_FUNC(svtbl2,_f64,,)(data, indices);
+}
More information about the cfe-commits
mailing list