[RISCV] Support Xsfvqmaccdod and Xsfvqmaccqoq extensions

clang/include/clang/Basic/riscv_sifive_vector.td

@@ -103,3 +103,27 @@ let SupportOverloading = false in {
     defm sf_vc_v_fvw : RVVVCIXBuiltinSet<["si"],  "UwKzUwUvFe", [-1, 0, 2, 3], UseGPR=0>;
   }
 }
+
+multiclass RVVVQMACCBuiltinSet<list<list<string>> suffixes_prototypes> {
+  let OverloadedName = NAME,
+      Name = NAME,
+      HasMasked = false,
+      Log2LMUL = [0, 1, 2, 3] in
+    defm NAME : RVVOutOp1Op2BuiltinSet<NAME, "i", suffixes_prototypes>;
+}
+
+let UnMaskedPolicyScheme = HasPolicyOperand in
+  let RequiredFeatures = ["Xsfvqmaccdod"] in {
+    defm sf_vqmaccu_2x8x2 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)SUv(FixedSEW:8)Uv"]]>;
+    defm sf_vqmacc_2x8x2 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)Sv(FixedSEW:8)v"]]>;
+    defm sf_vqmaccus_2x8x2 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)SUv(FixedSEW:8)v"]]>;
+    defm sf_vqmaccsu_2x8x2 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)Sv(FixedSEW:8)Uv"]]>;
+  }
+
+let UnMaskedPolicyScheme = HasPolicyOperand in
+  let RequiredFeatures = ["Xsfvqmaccqoq"] in {
+    defm sf_vqmaccu_4x8x4 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)SUv(FixedSEW:8)Uv"]]>;
+    defm sf_vqmacc_4x8x4 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)Sv(FixedSEW:8)v"]]>;
+    defm sf_vqmaccus_4x8x4 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)SUv(FixedSEW:8)v"]]>;
+    defm sf_vqmaccsu_4x8x4 : RVVVQMACCBuiltinSet<[["", "v", "vv(FixedSEW:8)Sv(FixedSEW:8)Uv"]]>;
+  }

clang/include/clang/Support/RISCVVIntrinsicUtils.h

@@ -485,14 +485,16 @@ enum RVVRequire : uint16_t {
   RVV_REQ_RV64 = 1 << 0,
   RVV_REQ_ZvfhminOrZvfh = 1 << 1,
   RVV_REQ_Xsfvcp = 1 << 2,
-  RVV_REQ_Zvbb = 1 << 3,
-  RVV_REQ_Zvbc = 1 << 4,
-  RVV_REQ_Zvkb = 1 << 5,
-  RVV_REQ_Zvkg = 1 << 6,
-  RVV_REQ_Zvkned = 1 << 7,
-  RVV_REQ_Zvknha = 1 << 8,
-  RVV_REQ_Zvksed = 1 << 9,
-  RVV_REQ_Zvksh = 1 << 10,
+  RVV_REQ_Xsfvqmaccdod = 1 << 3,
+  RVV_REQ_Xsfvqmaccqoq = 1 << 4,
+  RVV_REQ_Zvbb = 1 << 5,
+  RVV_REQ_Zvbc = 1 << 6,
+  RVV_REQ_Zvkb = 1 << 7,
+  RVV_REQ_Zvkg = 1 << 8,
+  RVV_REQ_Zvkned = 1 << 9,
+  RVV_REQ_Zvknha = 1 << 10,
+  RVV_REQ_Zvksed = 1 << 11,
+  RVV_REQ_Zvksh = 1 << 12,
 
   LLVM_MARK_AS_BITMASK_ENUM(RVV_REQ_Zvksh)
 };

clang/lib/Sema/SemaRISCVVectorLookup.cpp

@@ -205,6 +205,8 @@ void RISCVIntrinsicManagerImpl::ConstructRVVIntrinsics(
   static const std::pair<const char *, RVVRequire> FeatureCheckList[] = {
       {"64bit", RVV_REQ_RV64},
       {"xsfvcp", RVV_REQ_Xsfvcp},
+      {"xsfvqmaccdod", RVV_REQ_Xsfvqmaccdod},
+      {"xsfvqmaccqoq", RVV_REQ_Xsfvqmaccqoq},
       {"experimental-zvbb", RVV_REQ_Zvbb},
       {"experimental-zvbc", RVV_REQ_Zvbc},
       {"experimental-zvkb", RVV_REQ_Zvkb},

clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/sf_vqmacc_2x8x2.c

@@ -0,0 +1,47 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 2
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v -target-feature +xsfvqmaccdod \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <sifive_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i32> @test_sf_vqmacc_2x8x2_i32m1
+// CHECK-RV64-SAME: (<vscale x 2 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 8 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.sf.vqmacc.2x8x2.nxv2i32.nxv8i8.nxv8i8.i64(<vscale x 2 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 8 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_sf_vqmacc_2x8x2_i32m1(vint32m1_t vd, vint8m1_t vs1, vint8m1_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_2x8x2_i32m1(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i32> @test_sf_vqmacc_2x8x2_i32m2
+// CHECK-RV64-SAME: (<vscale x 4 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 16 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.sf.vqmacc.2x8x2.nxv4i32.nxv8i8.nxv16i8.i64(<vscale x 4 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 16 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_sf_vqmacc_2x8x2_i32m2(vint32m2_t vd, vint8m1_t vs1, vint8m2_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_2x8x2_i32m2(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i32> @test_sf_vqmacc_2x8x2_i32m4
+// CHECK-RV64-SAME: (<vscale x 8 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 32 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.sf.vqmacc.2x8x2.nxv8i32.nxv8i8.nxv32i8.i64(<vscale x 8 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 32 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_sf_vqmacc_2x8x2_i32m4(vint32m4_t vd, vint8m1_t vs1, vint8m4_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_2x8x2_i32m4(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vqmacc_2x8x2_i32m8
+// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 64 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vqmacc.2x8x2.nxv16i32.nxv8i8.nxv64i8.i64(<vscale x 16 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 64 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_sf_vqmacc_2x8x2_i32m8(vint32m8_t vd, vint8m1_t vs1, vint8m8_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_2x8x2_i32m8(vd, vs1, vs2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/sf_vqmacc_4x8x4.c

@@ -0,0 +1,47 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 2
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v -target-feature +xsfvqmaccqoq \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <sifive_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i32> @test_sf_vqmacc_4x8x4_i32m1
+// CHECK-RV64-SAME: (<vscale x 2 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 8 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.sf.vqmacc.4x8x4.nxv2i32.nxv8i8.nxv8i8.i64(<vscale x 2 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 8 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_sf_vqmacc_4x8x4_i32m1(vint32m1_t vd, vint8m1_t vs1, vint8m1_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_4x8x4_i32m1(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i32> @test_sf_vqmacc_4x8x4_i32m2
+// CHECK-RV64-SAME: (<vscale x 4 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 16 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.sf.vqmacc.4x8x4.nxv4i32.nxv8i8.nxv16i8.i64(<vscale x 4 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 16 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_sf_vqmacc_4x8x4_i32m2(vint32m2_t vd, vint8m1_t vs1, vint8m2_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_4x8x4_i32m2(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i32> @test_sf_vqmacc_4x8x4_i32m4
+// CHECK-RV64-SAME: (<vscale x 8 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 32 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.sf.vqmacc.4x8x4.nxv8i32.nxv8i8.nxv32i8.i64(<vscale x 8 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 32 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_sf_vqmacc_4x8x4_i32m4(vint32m4_t vd, vint8m1_t vs1, vint8m4_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_4x8x4_i32m4(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vqmacc_4x8x4_i32m8
+// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 64 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vqmacc.4x8x4.nxv16i32.nxv8i8.nxv64i8.i64(<vscale x 16 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 64 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_sf_vqmacc_4x8x4_i32m8(vint32m8_t vd, vint8m1_t vs1, vint8m8_t vs2, size_t vl) {
+  return __riscv_sf_vqmacc_4x8x4_i32m8(vd, vs1, vs2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/sf_vqmaccsu_2x8x2.c

@@ -0,0 +1,47 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 2
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v -target-feature +xsfvqmaccdod \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <sifive_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i32> @test_sf_vqmaccsu_2x8x2_i32m1
+// CHECK-RV64-SAME: (<vscale x 2 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 8 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.sf.vqmaccsu.2x8x2.nxv2i32.nxv8i8.nxv8i8.i64(<vscale x 2 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 8 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_sf_vqmaccsu_2x8x2_i32m1(vint32m1_t vd, vint8m1_t vs1, vuint8m1_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_2x8x2_i32m1(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i32> @test_sf_vqmaccsu_2x8x2_i32m2
+// CHECK-RV64-SAME: (<vscale x 4 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 16 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.sf.vqmaccsu.2x8x2.nxv4i32.nxv8i8.nxv16i8.i64(<vscale x 4 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 16 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_sf_vqmaccsu_2x8x2_i32m2(vint32m2_t vd, vint8m1_t vs1, vuint8m2_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_2x8x2_i32m2(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i32> @test_sf_vqmaccsu_2x8x2_i32m4
+// CHECK-RV64-SAME: (<vscale x 8 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 32 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.sf.vqmaccsu.2x8x2.nxv8i32.nxv8i8.nxv32i8.i64(<vscale x 8 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 32 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_sf_vqmaccsu_2x8x2_i32m4(vint32m4_t vd, vint8m1_t vs1, vuint8m4_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_2x8x2_i32m4(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vqmaccsu_2x8x2_i32m8
+// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 64 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vqmaccsu.2x8x2.nxv16i32.nxv8i8.nxv64i8.i64(<vscale x 16 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 64 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_sf_vqmaccsu_2x8x2_i32m8(vint32m8_t vd, vint8m1_t vs1, vuint8m8_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_2x8x2_i32m8(vd, vs1, vs2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/sf_vqmaccsu_4x8x4.c

@@ -0,0 +1,47 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 2
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v -target-feature +xsfvqmaccqoq \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <sifive_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i32> @test_sf_vqmaccsu_4x8x4_i32m1
+// CHECK-RV64-SAME: (<vscale x 2 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 8 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.sf.vqmaccsu.4x8x4.nxv2i32.nxv8i8.nxv8i8.i64(<vscale x 2 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 8 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_sf_vqmaccsu_4x8x4_i32m1(vint32m1_t vd, vint8m1_t vs1, vuint8m1_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_4x8x4_i32m1(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i32> @test_sf_vqmaccsu_4x8x4_i32m2
+// CHECK-RV64-SAME: (<vscale x 4 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 16 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.sf.vqmaccsu.4x8x4.nxv4i32.nxv8i8.nxv16i8.i64(<vscale x 4 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 16 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_sf_vqmaccsu_4x8x4_i32m2(vint32m2_t vd, vint8m1_t vs1, vuint8m2_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_4x8x4_i32m2(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i32> @test_sf_vqmaccsu_4x8x4_i32m4
+// CHECK-RV64-SAME: (<vscale x 8 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 32 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.sf.vqmaccsu.4x8x4.nxv8i32.nxv8i8.nxv32i8.i64(<vscale x 8 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 32 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_sf_vqmaccsu_4x8x4_i32m4(vint32m4_t vd, vint8m1_t vs1, vuint8m4_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_4x8x4_i32m4(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vqmaccsu_4x8x4_i32m8
+// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 64 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vqmaccsu.4x8x4.nxv16i32.nxv8i8.nxv64i8.i64(<vscale x 16 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 64 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_sf_vqmaccsu_4x8x4_i32m8(vint32m8_t vd, vint8m1_t vs1, vuint8m8_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccsu_4x8x4_i32m8(vd, vs1, vs2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/sf_vqmaccu_2x8x2.c

@@ -0,0 +1,47 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 2
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v -target-feature +xsfvqmaccdod \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <sifive_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i32> @test_sf_vqmaccu_2x8x2_i32m1
+// CHECK-RV64-SAME: (<vscale x 2 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 8 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.sf.vqmaccu.2x8x2.nxv2i32.nxv8i8.nxv8i8.i64(<vscale x 2 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 8 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_sf_vqmaccu_2x8x2_i32m1(vint32m1_t vd, vuint8m1_t vs1, vuint8m1_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccu_2x8x2_i32m1(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i32> @test_sf_vqmaccu_2x8x2_i32m2
+// CHECK-RV64-SAME: (<vscale x 4 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 16 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.sf.vqmaccu.2x8x2.nxv4i32.nxv8i8.nxv16i8.i64(<vscale x 4 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 16 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_sf_vqmaccu_2x8x2_i32m2(vint32m2_t vd, vuint8m1_t vs1, vuint8m2_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccu_2x8x2_i32m2(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i32> @test_sf_vqmaccu_2x8x2_i32m4
+// CHECK-RV64-SAME: (<vscale x 8 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 32 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.sf.vqmaccu.2x8x2.nxv8i32.nxv8i8.nxv32i8.i64(<vscale x 8 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 32 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_sf_vqmaccu_2x8x2_i32m4(vint32m4_t vd, vuint8m1_t vs1, vuint8m4_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccu_2x8x2_i32m4(vd, vs1, vs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vqmaccu_2x8x2_i32m8
+// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 8 x i8> [[VS1:%.*]], <vscale x 64 x i8> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vqmaccu.2x8x2.nxv16i32.nxv8i8.nxv64i8.i64(<vscale x 16 x i32> [[VD]], <vscale x 8 x i8> [[VS1]], <vscale x 64 x i8> [[VS2]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_sf_vqmaccu_2x8x2_i32m8(vint32m8_t vd, vuint8m1_t vs1, vuint8m8_t vs2, size_t vl) {
+  return __riscv_sf_vqmaccu_2x8x2_i32m8(vd, vs1, vs2, vl);
+}