[AArch64][SME2] Add SQRSHRN, UQRSHRN, SQRSHRUN builtins for SME2, SVE2p1 #75325
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add SQRSHRN, UQRSHRN, SQRSHRUN builtins for SME2, SVE2p1.
dtemirbulatov
requested review from
SamTebbs33,
sdesmalen-arm,
CarolineConcatto and
kmclaughlin-arm
llvmbot
added
clang
|
@llvm/pr-subscribers-clang Author: Dinar Temirbulatov (dtemirbulatov) ChangesAdd SQRSHRN, UQRSHRN, SQRSHRUN builtins for SME2, SVE2p1. Full diff: https://github.com/llvm/llvm-project/pull/75325.diff 2 Files Affected:
diff --git a/clang/include/clang/Basic/arm_sve.td b/clang/include/clang/Basic/arm_sve.td
index aa9b105364a51a..4e9e318e7ade85 100644
--- a/clang/include/clang/Basic/arm_sve.td
+++ b/clang/include/clang/Basic/arm_sve.td
@@ -2170,6 +2170,15 @@ let TargetGuard = "sme2" in {
def REINTERPRET_SVCOUNT_TO_SVBOOL : Inst<"svreinterpret[_b]", "P}", "Pc", MergeNone, "", [IsStreamingCompatible], []>;
}
+let TargetGuard = "sve2p1|sme2" in {
+ // SQRSHRN / UQRSHRN
+ def SVQRSHRN_X2 : SInst<"svqrshrn[_n]_{0}[_{d}_x2]", "h2i", "i", MergeNone, "aarch64_sve_sqrshrn_x2", [IsStreamingCompatible], [ImmCheck<1, ImmCheck1_16>]>;
+ def SVUQRSHRN_X2 : SInst<"svqrshrn[_n]_{0}[_{d}_x2]", "e2i", "Ui", MergeNone, "aarch64_sve_uqrshrn_x2", [IsStreamingCompatible], [ImmCheck<1, ImmCheck1_16>]>;
+
+ // SQRSHRUN
+ def SVSQRSHRUN_X2 : SInst<"svqrshrun[_n]_{0}[_{d}_x2]", "e2i", "i", MergeNone, "aarch64_sve_sqrshrun_x2", [IsStreamingCompatible], [ImmCheck<1, ImmCheck1_16>]>;
+}
+
let TargetGuard = "sve2p1" in {
// ZIPQ1, ZIPQ2, UZPQ1, UZPQ2
def SVZIPQ1 : SInst<"svzipq1[_{d}]", "ddd", "cUcsUsiUilUlbhfd", MergeNone, "aarch64_sve_zipq1", [], []>;
diff --git a/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_qrshr.c b/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_qrshr.c
new file mode 100644
index 00000000000000..8e8b7203148934
--- /dev/null
+++ b/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_qrshr.c
@@ -0,0 +1,78 @@
+// 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 +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve2p1 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
+// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -target-feature +sme-f64f64 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve2p1 -S -disable-O0-optnone -Werror -Wall -o /dev/null %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,A5) A1##A3##A5
+#else
+#define SVE_ACLE_FUNC(A1,A2,A3,A4,A5) A1##A2##A3##A4##A5
+#endif
+
+
+// SQRSHRN x 2
+
+// CHECK-LABEL: @test_svqrshrn_s16_s32_x2(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN:%.*]], i64 0)
+// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN]], i64 4)
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.sqrshrn.x2.nxv4i32(<vscale x 4 x i32> [[TMP0]], <vscale x 4 x i32> [[TMP1]], i32 16)
+// CHECK-NEXT: ret <vscale x 8 x i16> [[TMP2]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svqrshrn_s16_s32_x211svint32x2_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN:%.*]], i64 0)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN]], i64 4)
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.sqrshrn.x2.nxv4i32(<vscale x 4 x i32> [[TMP0]], <vscale x 4 x i32> [[TMP1]], i32 16)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP2]]
+//
+svint16_t test_svqrshrn_s16_s32_x2(svint32x2_t zn) __arm_streaming_compatible {
+ return SVE_ACLE_FUNC(svqrshrn,_n,_s16,_s32_x2,)(zn, 16);
+}
+
+// UQRSHRN x 2
+
+// CHECK-LABEL: @test_svqrshrn_u16_u32_x2(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN:%.*]], i64 0)
+// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN]], i64 4)
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.uqrshrn.x2.nxv4i32(<vscale x 4 x i32> [[TMP0]], <vscale x 4 x i32> [[TMP1]], i32 16)
+// CHECK-NEXT: ret <vscale x 8 x i16> [[TMP2]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svqrshrn_u16_u32_x212svuint32x2_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN:%.*]], i64 0)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN]], i64 4)
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.uqrshrn.x2.nxv4i32(<vscale x 4 x i32> [[TMP0]], <vscale x 4 x i32> [[TMP1]], i32 16)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP2]]
+//
+svuint16_t test_svqrshrn_u16_u32_x2(svuint32x2_t zn) __arm_streaming_compatible {
+ return SVE_ACLE_FUNC(svqrshrn,_n,_u16,_u32_x2,)(zn, 16);
+}
+
+// SQRSHRUN x 2
+
+// CHECK-LABEL: @test_svqrshrun_u16_s32_x2(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN:%.*]], i64 0)
+// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN]], i64 4)
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.sqrshrun.x2.nxv4i32(<vscale x 4 x i32> [[TMP0]], <vscale x 4 x i32> [[TMP1]], i32 16)
+// CHECK-NEXT: ret <vscale x 8 x i16> [[TMP2]]
+//
+// CPP-CHECK-LABEL: @_Z25test_svqrshrun_u16_s32_x211svint32x2_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN:%.*]], i64 0)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> [[ZN]], i64 4)
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.sqrshrun.x2.nxv4i32(<vscale x 4 x i32> [[TMP0]], <vscale x 4 x i32> [[TMP1]], i32 16)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP2]]
+//
+svuint16_t test_svqrshrun_u16_s32_x2(svint32x2_t zn) __arm_streaming_compatible {
+ return SVE_ACLE_FUNC(svqrshrun,_n,_u16,_s32_x2,)(zn, 16);
+}
|
CarolineConcatto
approved these changes
Dec 13, 2023
|
Please remember to add a suitable This new test is failing in non-ARM builds with: |
I've added one in 50e78de |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add SQRSHRN, UQRSHRN, SQRSHRUN builtins for SME2, SVE2p1.