[SVE ACLE] Change the lowering of SVE integer builtins

[jolanta.jensen]

Summary

Change the lowering of SVE integer mla_x/mls_x and mad_x/msb_x
builtins to use dedicated undef (_u) intrinsics.

[mgabka]

llvm/include/llvm/IR/IntrinsicsAArch64.td

1666

it is not a bug, but just a preference to keep better ordering here, could you move this new definition to be straight below the def int_aarch64_sve_mla ? so the regular and _u intrinsics are defined in the same order as in all other cases.

Same request to the mls_u definition.

[jolanta.jensen]

Minor reordering after review comment.

[paulwalker-arm]

The commit message is incorrect because the SVE Clang builtins are not named mla_u and so should read SVE integer mla_x/mls_x builtins Other than that the patch looks good to me with the following suggestions.

llvm/test/CodeGen/AArch64/sve-intrinsics-int-arith-undef.ll

1307

Just because this is a big file can you follow the existing scheme of sorting alphabetically (i.e. move the new block just before the @mul functions and prefix the two blocks with

;
; MLA
;

;
; MLS
;

accordingly. This just makes it easier to find things when manually scrolling.

1379–1387

As above, please maintain the ordering by moving these declaration just before those for mul.

[jolanta.jensen]

Corrected the commit message.
Added tests are now in alphabetical order.

clang/include/clang/Basic/arm_sve.td

multiclass SInstZPZZZ<string name, string types, string m_intrinsic, string x_intrinsic, list<FlagType> flags=[]> {
   def _X   : SInst<name # "[_{d}]",   "dPddd", types, MergeAny,  x_intrinsic, flags>;
   def _Z   : SInst<name # "[_{d}]",   "dPddd", types, MergeZero, m_intrinsic, flags>;
 
   def _N_M : SInst<name # "[_n_{d}]", "dPdda", types, MergeOp1,  m_intrinsic, flags>;
   def _N_X : SInst<name # "[_n_{d}]", "dPdda", types, MergeAny,  x_intrinsic, flags>;
   def _N_Z : SInst<name # "[_n_{d}]", "dPdda", types, MergeZero, m_intrinsic, flags>;
 }
 
-defm SVMAD : SInstZPZZZ<"svmad", "csilUcUsUiUl", "aarch64_sve_mad", "aarch64_sve_mad">;
-defm SVMLA : SInstZPZZZ<"svmla", "csilUcUsUiUl", "aarch64_sve_mla", "aarch64_sve_mla">;
-defm SVMLS : SInstZPZZZ<"svmls", "csilUcUsUiUl", "aarch64_sve_mls", "aarch64_sve_mls">;
-defm SVMSB : SInstZPZZZ<"svmsb", "csilUcUsUiUl", "aarch64_sve_msb", "aarch64_sve_msb">;
+defm SVMAD : SInstZPZZZ<"svmad", "csilUcUsUiUl", "aarch64_sve_mad", "aarch64_sve_mla_u", [ReverseMergeAnyAccOp]>;
+defm SVMLA : SInstZPZZZ<"svmla", "csilUcUsUiUl", "aarch64_sve_mla", "aarch64_sve_mla_u">;
+defm SVMLS : SInstZPZZZ<"svmls", "csilUcUsUiUl", "aarch64_sve_mls", "aarch64_sve_mls_u">;
+defm SVMSB : SInstZPZZZ<"svmsb", "csilUcUsUiUl", "aarch64_sve_msb", "aarch64_sve_mls_u", [ReverseMergeAnyAccOp]>;
 
 //------------------------------------------------------------------------------
 
 def SVDOT_S    : SInst<"svdot[_{0}]",    "ddqq", "il",       MergeNone, "aarch64_sve_sdot">;
 def SVDOT_U    : SInst<"svdot[_{0}]",    "ddqq", "UiUl",     MergeNone, "aarch64_sve_udot">;
 def SVQADD_S   : SInst<"svqadd[_{d}]",   "ddd",  "csil",     MergeNone, "aarch64_sve_sqadd_x">;
 def SVQADD_U   : SInst<"svqadd[_{d}]",   "ddd",  "UcUsUiUl", MergeNone, "aarch64_sve_uqadd_x">;
 def SVQSUB_S   : SInst<"svqsub[_{d}]",   "ddd",  "csil",     MergeNone, "aarch64_sve_sqsub_x">;

clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_mad.c

 //
 svuint64_t test_svmad_u64_m(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_s8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmad_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_tu10__SVInt8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmad_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, svint8_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmad_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_tu11__SVInt16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmad_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, svint16_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmad_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_tu11__SVInt32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmad_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, svint32_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmad_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tu11__SVInt64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmad_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, svint64_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_u8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmad_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_tu11__SVUint8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmad_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, svuint8_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmad_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tu12__SVUint16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmad_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, svuint16_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmad_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tu12__SVUint32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmad_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, svuint32_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmad_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tu12__SVUint64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmad_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_s8_z(

 {
   return SVE_ACLE_FUNC(svmad,_n_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_s8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmad_n_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_ta(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmad_n_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, int8_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmad_n_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_ts(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmad_n_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, int16_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmad_n_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_ti(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmad_n_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, int32_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmad_n_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tl(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmad_n_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, int64_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_u8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmad_n_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_th(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmad_n_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, uint8_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmad_n_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tt(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mad.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmad_n_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, uint16_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmad_n_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tj(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mad.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmad_n_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, uint32_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_n_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmad_n_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tm(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mad.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmad_n_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, uint64_t op3)
 {
   return SVE_ACLE_FUNC(svmad,_n_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmad_f16_z(

clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_mla.c

 //
 svuint64_t test_svmla_u64_m(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_s8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmla_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_tu10__SVInt8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmla_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, svint8_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmla_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_tu11__SVInt16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmla_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, svint16_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmla_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_tu11__SVInt32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmla_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, svint32_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmla_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tu11__SVInt64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmla_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, svint64_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_u8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmla_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_tu11__SVUint8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmla_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, svuint8_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmla_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tu12__SVUint16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmla_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, svuint16_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmla_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tu12__SVUint32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmla_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, svuint32_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmla_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tu12__SVUint64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmla_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_s8_z(

 {
   return SVE_ACLE_FUNC(svmla,_n_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_s8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmla_n_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_ta(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmla_n_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, int8_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmla_n_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_ts(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmla_n_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, int16_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmla_n_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_ti(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmla_n_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, int32_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmla_n_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tl(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmla_n_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, int64_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_u8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmla_n_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_th(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmla_n_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, uint8_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmla_n_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tt(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmla_n_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, uint16_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmla_n_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tj(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmla_n_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, uint32_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_n_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmla_n_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tm(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmla_n_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, uint64_t op3)
 {
   return SVE_ACLE_FUNC(svmla,_n_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmla_f16_z(

clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_mls.c

 //
 svuint64_t test_svmls_u64_m(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_s8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmls_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_tu10__SVInt8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmls_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, svint8_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmls_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_tu11__SVInt16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmls_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, svint16_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmls_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_tu11__SVInt32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmls_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, svint32_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmls_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tu11__SVInt64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmls_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, svint64_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_u8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmls_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_tu11__SVUint8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmls_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, svuint8_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmls_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tu12__SVUint16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmls_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, svuint16_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmls_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tu12__SVUint32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmls_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, svuint32_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmls_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tu12__SVUint64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmls_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_s8_z(

 {
   return SVE_ACLE_FUNC(svmls,_n_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_s8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmls_n_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_ta(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmls_n_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, int8_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmls_n_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_ts(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmls_n_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, int16_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmls_n_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_ti(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmls_n_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, int32_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmls_n_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tl(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmls_n_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, int64_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_u8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmls_n_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_th(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmls_n_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, uint8_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmls_n_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tt(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmls_n_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, uint16_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmls_n_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tj(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmls_n_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, uint32_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_n_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmls_n_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tm(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmls_n_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, uint64_t op3)
 {
   return SVE_ACLE_FUNC(svmls,_n_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmls_f16_z(

clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_msb.c

 //
 svuint64_t test_svmsb_u64_m(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_s8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmsb_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_tu10__SVInt8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmsb_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, svint8_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmsb_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_tu11__SVInt16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmsb_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, svint16_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmsb_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_tu11__SVInt32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmsb_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, svint32_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmsb_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tu11__SVInt64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmsb_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, svint64_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_u8_x(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z15test_svmsb_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_tu11__SVUint8_t(
 // CPP-CHECK-NEXT:  entry:
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP3:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmsb_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, svuint8_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmsb_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tu12__SVUint16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP3:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmsb_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, svuint16_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmsb_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tu12__SVUint32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP3:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmsb_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, svuint32_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svmsb_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tu12__SVUint64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP3:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmsb_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, svuint64_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_s8_z(

 {
   return SVE_ACLE_FUNC(svmsb,_n_u64,_m,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_s8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmsb_n_s8_xu10__SVBool_tu10__SVInt8_tu10__SVInt8_ta(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svint8_t test_svmsb_n_s8_x(svbool_t pg, svint8_t op1, svint8_t op2, int8_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_s8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_s16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmsb_n_s16_xu10__SVBool_tu11__SVInt16_tu11__SVInt16_ts(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svint16_t test_svmsb_n_s16_x(svbool_t pg, svint16_t op1, svint16_t op2, int16_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_s16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_s32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmsb_n_s32_xu10__SVBool_tu11__SVInt32_tu11__SVInt32_ti(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svint32_t test_svmsb_n_s32_x(svbool_t pg, svint32_t op1, svint32_t op2, int32_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_s32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_s64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmsb_n_s64_xu10__SVBool_tu11__SVInt64_tu11__SVInt64_tl(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svint64_t test_svmsb_n_s64_x(svbool_t pg, svint64_t op1, svint64_t op2, int64_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_s64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_u8_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 // CPP-CHECK-LABEL: @_Z17test_svmsb_n_u8_xu10__SVBool_tu11__SVUint8_tu11__SVUint8_th(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 16 x i8> poison, i8 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 16 x i8> [[DOTSPLATINSERT]], <vscale x 16 x i8> poison, <vscale x 16 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.msb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[DOTSPLAT]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 svuint8_t test_svmsb_n_u8_x(svbool_t pg, svuint8_t op1, svuint8_t op2, uint8_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_u8,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_u16_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmsb_n_u16_xu10__SVBool_tu12__SVUint16_tu12__SVUint16_tt(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[DOTSPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.msb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[DOTSPLAT]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP1]]
 //
 svuint16_t test_svmsb_n_u16_x(svbool_t pg, svuint16_t op1, svuint16_t op2, uint16_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_u16,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_u32_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmsb_n_u32_xu10__SVBool_tu12__SVUint32_tu12__SVUint32_tj(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[DOTSPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.msb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[DOTSPLAT]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
 //
 svuint32_t test_svmsb_n_u32_x(svbool_t pg, svuint32_t op1, svuint32_t op2, uint32_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_u32,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_n_u64_x(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z18test_svmsb_n_u64_xu10__SVBool_tu12__SVUint64_tu12__SVUint64_tm(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
 // CPP-CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[OP3:%.*]], i64 0
 // CPP-CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.msb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[DOTSPLAT]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[DOTSPLAT]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP1:%.*]])
 // CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
 //
 svuint64_t test_svmsb_n_u64_x(svbool_t pg, svuint64_t op1, svuint64_t op2, uint64_t op3)
 {
   return SVE_ACLE_FUNC(svmsb,_n_u64,_x,)(pg, op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svmsb_f16_z(

llvm/include/llvm/IR/IntrinsicsAArch64.td

 def int_aarch64_sve_sabd       : AdvSIMD_Pred2VectorArg_Intrinsic;
 def int_aarch64_sve_sabd_u     : AdvSIMD_Pred2VectorArg_Intrinsic;
 def int_aarch64_sve_uabd       : AdvSIMD_Pred2VectorArg_Intrinsic;
 def int_aarch64_sve_uabd_u     : AdvSIMD_Pred2VectorArg_Intrinsic;
 
 def int_aarch64_sve_mad        : AdvSIMD_Pred3VectorArg_Intrinsic;
 def int_aarch64_sve_msb        : AdvSIMD_Pred3VectorArg_Intrinsic;
 def int_aarch64_sve_mla        : AdvSIMD_Pred3VectorArg_Intrinsic;
+def int_aarch64_sve_mla_u      : AdvSIMD_Pred3VectorArg_Intrinsic;
 def int_aarch64_sve_mla_lane   : AdvSIMD_3VectorArgIndexed_Intrinsic;
 def int_aarch64_sve_mls        : AdvSIMD_Pred3VectorArg_Intrinsic;
+def int_aarch64_sve_mls_u      : AdvSIMD_Pred3VectorArg_Intrinsic;
 def int_aarch64_sve_mls_lane   : AdvSIMD_3VectorArgIndexed_Intrinsic;
 
 def int_aarch64_sve_saddv      : AdvSIMD_SVE_SADDV_Reduce_Intrinsic;
 def int_aarch64_sve_uaddv      : AdvSIMD_SVE_SADDV_Reduce_Intrinsic;
 
 def int_aarch64_sve_smaxv      : AdvSIMD_SVE_Reduce_Intrinsic;
 def int_aarch64_sve_umaxv      : AdvSIMD_SVE_Reduce_Intrinsic;
 def int_aarch64_sve_sminv      : AdvSIMD_SVE_Reduce_Intrinsic;

llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td

def AArch64sub_m1 : PatFrags<(ops node:$pred, node:$op1, node:$op2),
                              [(int_aarch64_sve_sub node:$pred, node:$op1, node:$op2),
                               (sub node:$op1, (vselect node:$pred, node:$op2, (SVEDup0)))]>;
 def AArch64mla_m1 : PatFrags<(ops node:$pred, node:$op1, node:$op2, node:$op3),
                              [(int_aarch64_sve_mla node:$pred, node:$op1, node:$op2, node:$op3),
                               // add(a, select(mask, mul(b, c), splat(0))) -> mla(a, mask, b, c)
                               (add node:$op1, (vselect node:$pred, (AArch64mul_p_oneuse (SVEAllActive), node:$op2, node:$op3), (SVEDup0)))]>;
 // pattern for generating pseudo for MLA_ZPmZZ/MAD_ZPmZZ
 def AArch64mla_p : PatFrags<(ops node:$pred, node:$op1, node:$op2, node:$op3),
-                            [(add node:$op1, (AArch64mul_p_oneuse node:$pred, node:$op2, node:$op3))]>;
+                            [(int_aarch64_sve_mla_u node:$pred, node:$op1, node:$op2, node:$op3),
+                             (add node:$op1, (AArch64mul_p_oneuse node:$pred, node:$op2, node:$op3))]>;
 def AArch64mls_m1 : PatFrags<(ops node:$pred, node:$op1, node:$op2, node:$op3),
                              [(int_aarch64_sve_mls node:$pred, node:$op1, node:$op2, node:$op3),
                               // sub(a, select(mask, mul(b, c), splat(0))) -> mls(a, mask, b, c)
                               (sub node:$op1, (vselect node:$pred, (AArch64mul_p_oneuse (SVEAllActive), node:$op2, node:$op3), (SVEDup0)))]>;
 def AArch64mls_p : PatFrags<(ops node:$pred, node:$op1, node:$op2, node:$op3),
-                             [(sub node:$op1, (AArch64mul_p_oneuse node:$pred, node:$op2, node:$op3))]>;
+                             [(int_aarch64_sve_mls_u node:$pred, node:$op1, node:$op2, node:$op3),
+                              (sub node:$op1, (AArch64mul_p_oneuse node:$pred, node:$op2, node:$op3))]>;
 def AArch64eor3 : PatFrags<(ops node:$op1, node:$op2, node:$op3),
                            [(int_aarch64_sve_eor3 node:$op1, node:$op2, node:$op3),
                             (xor node:$op1, (xor node:$op2, node:$op3))]>;
 
 class fma_patfrags<SDPatternOperator intrinsic, SDPatternOperator add_zero, SDPatternOperator add_negzero>
     : PatFrags<(ops node:$pred, node:$op1, node:$op2, node:$op3),
                [(intrinsic node:$pred, node:$op1, node:$op2, node:$op3),
                 (add_zero (SVEAllActive), node:$op1, (vselect node:$pred, (AArch64fmul_p_oneuse (SVEAllActive), node:$op2, node:$op3), (SVEDup0))),

llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp

case Intrinsic::aarch64_sve_fmul:
     return instCombineSVEVectorMul(IC, II);
   case Intrinsic::aarch64_sve_fadd:
   case Intrinsic::aarch64_sve_add:
     return instCombineSVEVectorAdd(IC, II);
   case Intrinsic::aarch64_sve_fadd_u:
     return instCombineSVEVectorFuseMulAddSub<Intrinsic::aarch64_sve_fmul_u,
                                              Intrinsic::aarch64_sve_fmla_u>(
         IC, II, true);
+  case Intrinsic::aarch64_sve_add_u:
+    return instCombineSVEVectorFuseMulAddSub<Intrinsic::aarch64_sve_mul_u,
+                                             Intrinsic::aarch64_sve_mla_u>(
+        IC, II, true);
   case Intrinsic::aarch64_sve_fsub:
   case Intrinsic::aarch64_sve_sub:
     return instCombineSVEVectorSub(IC, II);
   case Intrinsic::aarch64_sve_fsub_u:
     return instCombineSVEVectorFuseMulAddSub<Intrinsic::aarch64_sve_fmul_u,
                                              Intrinsic::aarch64_sve_fmls_u>(
         IC, II, true);
+  case Intrinsic::aarch64_sve_sub_u:
+    return instCombineSVEVectorFuseMulAddSub<Intrinsic::aarch64_sve_mul_u,
+                                             Intrinsic::aarch64_sve_mls_u>(
+        IC, II, true);
   case Intrinsic::aarch64_sve_tbl:
     return instCombineSVETBL(IC, II);
   case Intrinsic::aarch64_sve_uunpkhi:
   case Intrinsic::aarch64_sve_uunpklo:
   case Intrinsic::aarch64_sve_sunpkhi:
   case Intrinsic::aarch64_sve_sunpklo:
     return instCombineSVEUnpack(IC, II);
   case Intrinsic::aarch64_sve_zip1:

llvm/test/CodeGen/AArch64/sve-intrinsics-int-arith-undef.ll

; CHECK-NEXT:    ret
   %imm.splat = shufflevector <vscale x 2 x i64> %imm, <vscale x 2 x i64> undef, <vscale x 2 x i32> zeroinitializer
   %out = call <vscale x 2 x i64> @llvm.aarch64.sve.add.u.nxv2i64(<vscale x 2 x i1> %pg,
                                                                  <vscale x 2 x i64> %a,
                                                                  <vscale x 2 x i64> %imm.splat)
   ret <vscale x 2 x i64> %out
 }
 
 ;
+; MLA
+;
+
+define <vscale x 16 x i8> @mla_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) {
+; CHECK-LABEL: mla_i8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mla z0.b, p0/m, z1.b, z2.b
+; CHECK-NEXT:    ret
+  %out = call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> %pg,
+                                                                 <vscale x 16 x i8> %a,
+                                                                 <vscale x 16 x i8> %b,
+                                                                 <vscale x 16 x i8> %c)
+  ret <vscale x 16 x i8> %out
+}
+
+define <vscale x 8 x i16> @mla_i16(<vscale x 8 x i1> %pg, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b, <vscale x 8 x i16> %c) {
+; CHECK-LABEL: mla_i16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mla z0.h, p0/m, z1.h, z2.h
+; CHECK-NEXT:    ret
+  %out = call <vscale x 8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x 8 x i1> %pg,
+                                                                 <vscale x 8 x i16> %a,
+                                                                 <vscale x 8 x i16> %b,
+                                                                 <vscale x 8 x i16> %c)
+  ret <vscale x 8 x i16> %out
+}
+
+define <vscale x 4 x i32> @mla_i32(<vscale x 4 x i1> %pg, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
+; CHECK-LABEL: mla_i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mla z0.s, p0/m, z1.s, z2.s
+; CHECK-NEXT:    ret
+  %out = call <vscale x 4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x 4 x i1> %pg,
+                                                                 <vscale x 4 x i32> %a,
+                                                                 <vscale x 4 x i32> %b,
+                                                                 <vscale x 4 x i32> %c)
+  ret <vscale x 4 x i32> %out
+}
+
+define <vscale x 2 x i64> @mla_i64(<vscale x 2 x i1> %pg, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b, <vscale x 2 x i64> %c) {
+; CHECK-LABEL: mla_i64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mla z0.d, p0/m, z1.d, z2.d
+; CHECK-NEXT:    ret
+  %out = call <vscale x 2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x 2 x i1> %pg,
+                                                                 <vscale x 2 x i64> %a,
+                                                                 <vscale x 2 x i64> %b,
+                                                                 <vscale x 2 x i64> %c)
+  ret <vscale x 2 x i64> %out
+}
+
+;
+; MLS
+;
+
+define <vscale x 16 x i8> @mls_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) {
+; CHECK-LABEL: mls_i8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mls z0.b, p0/m, z1.b, z2.b
+; CHECK-NEXT:    ret
+  %out = call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> %pg,
+                                                                 <vscale x 16 x i8> %a,
+                                                                 <vscale x 16 x i8> %b,
+                                                                 <vscale x 16 x i8> %c)
+  ret <vscale x 16 x i8> %out
+}
+
+define <vscale x 8 x i16> @mls_i16(<vscale x 8 x i1> %pg, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b, <vscale x 8 x i16> %c) {
+; CHECK-LABEL: mls_i16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mls z0.h, p0/m, z1.h, z2.h
+; CHECK-NEXT:    ret
+  %out = call <vscale x 8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x 8 x i1> %pg,
+                                                                 <vscale x 8 x i16> %a,
+                                                                 <vscale x 8 x i16> %b,
+                                                                 <vscale x 8 x i16> %c)
+  ret <vscale x 8 x i16> %out
+}
+
+define <vscale x 4 x i32> @mls_i32(<vscale x 4 x i1> %pg, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
+; CHECK-LABEL: mls_i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mls z0.s, p0/m, z1.s, z2.s
+; CHECK-NEXT:    ret
+  %out = call <vscale x 4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x 4 x i1> %pg,
+                                                                 <vscale x 4 x i32> %a,
+                                                                 <vscale x 4 x i32> %b,
+                                                                 <vscale x 4 x i32> %c)
+  ret <vscale x 4 x i32> %out
+}
+
+define <vscale x 2 x i64> @mls_i64(<vscale x 2 x i1> %pg, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b, <vscale x 2 x i64> %c) {
+; CHECK-LABEL: mls_i64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mls z0.d, p0/m, z1.d, z2.d
+; CHECK-NEXT:    ret
+  %out = call <vscale x 2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x 2 x i1> %pg,
+                                                                 <vscale x 2 x i64> %a,
+                                                                 <vscale x 2 x i64> %b,
+                                                                 <vscale x 2 x i64> %c)
+  ret <vscale x 2 x i64> %out
+}
+
+;
 ; MUL
 ;
 
 define <vscale x 16 x i8> @mul_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
 ; SVE1-LABEL: mul_i8:
 ; SVE1:       // %bb.0:
 ; SVE1-NEXT:    mul z0.b, p0/m, z0.b, z1.b
 ; SVE1-NEXT:    ret

 ; SVE2-NEXT:    umulh z0.d, z0.d, z1.d
 ; SVE2-NEXT:    ret
   %out = call <vscale x 2 x i64> @llvm.aarch64.sve.umulh.u.nxv2i64(<vscale x 2 x i1> %pg,
                                                                    <vscale x 2 x i64> %a,
                                                                    <vscale x 2 x i64> %b)
   ret <vscale x 2 x i64> %out
 }
 
+
 declare <vscale x 16 x  i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
 declare <vscale x  8 x i16> @llvm.aarch64.sve.add.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>)
 declare <vscale x  4 x i32> @llvm.aarch64.sve.add.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>)
 declare <vscale x  2 x i64> @llvm.aarch64.sve.add.u.nxv2i64(<vscale x  2 x i1>, <vscale x  2 x i64>, <vscale x  2 x i64>)
 
+declare <vscale x 16 x  i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
+declare <vscale x  8 x i16> @llvm.aarch64.sve.mla.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>, <vscale x  8 x i16>)
+declare <vscale x  4 x i32> @llvm.aarch64.sve.mla.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>, <vscale x  4 x i32>)
+declare <vscale x  2 x i64> @llvm.aarch64.sve.mla.u.nxv2i64(<vscale x  2 x i1>, <vscale x  2 x i64>, <vscale x  2 x i64>, <vscale x  2 x i64>)
+
+declare <vscale x 16 x  i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
+declare <vscale x  8 x i16> @llvm.aarch64.sve.mls.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>, <vscale x  8 x i16>)
+declare <vscale x  4 x i32> @llvm.aarch64.sve.mls.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>, <vscale x  4 x i32>)
+declare <vscale x  2 x i64> @llvm.aarch64.sve.mls.u.nxv2i64(<vscale x  2 x i1>, <vscale x  2 x i64>, <vscale x  2 x i64>, <vscale x  2 x i64>)
+
 declare <vscale x 16 x  i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
 declare <vscale x  8 x i16> @llvm.aarch64.sve.mul.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>)
 declare <vscale x  4 x i32> @llvm.aarch64.sve.mul.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>)
 declare <vscale x  2 x i64> @llvm.aarch64.sve.mul.u.nxv2i64(<vscale x  2 x i1>, <vscale x  2 x i64>, <vscale x  2 x i64>)
 
 declare <vscale x 16 x  i8> @llvm.aarch64.sve.sabd.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
 declare <vscale x  8 x i16> @llvm.aarch64.sve.sabd.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>)
 declare <vscale x  4 x i32> @llvm.aarch64.sve.sabd.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>)

 declare <vscale x 16 x  i8> @llvm.aarch64.sve.umin.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
 declare <vscale x  8 x i16> @llvm.aarch64.sve.umin.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>)
 declare <vscale x  4 x i32> @llvm.aarch64.sve.umin.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>)
 declare <vscale x  2 x i64> @llvm.aarch64.sve.umin.u.nxv2i64(<vscale x  2 x i1>, <vscale x  2 x i64>, <vscale x  2 x i64>)
 
 declare <vscale x 16 x  i8> @llvm.aarch64.sve.umulh.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x  i8>, <vscale x 16 x  i8>)
 declare <vscale x  8 x i16> @llvm.aarch64.sve.umulh.u.nxv8i16(<vscale x  8 x i1>, <vscale x  8 x i16>, <vscale x  8 x i16>)
 declare <vscale x  4 x i32> @llvm.aarch64.sve.umulh.u.nxv4i32(<vscale x  4 x i1>, <vscale x  4 x i32>, <vscale x  4 x i32>)
 declare <vscale x  2 x i64> @llvm.aarch64.sve.umulh.u.nxv2i64(<vscale x  2 x i1>, <vscale x  2 x i64>, <vscale x  2 x i64>)

llvm/test/Transforms/InstCombine/AArch64/sve-intrinsic-muladdsub.ll

 ; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[C:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]])
 ; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
 ;
   %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
   %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %c, <vscale x 16 x i8> %1)
   ret <vscale x 16 x i8> %2
 }
 
+define <vscale x 16 x i8> @combine_mla_i8_u(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
+; CHECK-LABEL: @combine_mla_i8_u(
+; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[C:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]])
+; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
+;
+  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
+  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %c, <vscale x 16 x i8> %1)
+  ret <vscale x 16 x i8> %2
+}
+
 define <vscale x 8 x half> @combine_fmad(<vscale x 16 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
 ; CHECK-LABEL: @combine_fmad(
 ; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[P:%.*]])
 ; CHECK-NEXT:    [[TMP2:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmad.nxv8f16(<vscale x 8 x i1> [[TMP1]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
 ; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
 ;
   %1 = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> %p)
   %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %1, <vscale x 8 x half> %a, <vscale x 8 x half> %b)

 ; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[C:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]])
 ; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
 ;
   %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
   %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %c, <vscale x 16 x i8> %1)
   ret <vscale x 16 x i8> %2
 }
 
+define <vscale x 16 x i8> @combine_mls_i8_u(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
+; CHECK-LABEL: @combine_mls_i8_u(
+; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[C:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]])
+; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
+;
+  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
+  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %c, <vscale x 16 x i8> %1)
+  ret <vscale x 16 x i8> %2
+}
+
 define <vscale x 8 x half> @combine_fnmsb(<vscale x 16 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
 ; CHECK-LABEL: @combine_fnmsb(
 ; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[P:%.*]])
 ; CHECK-NEXT:    [[TMP2:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fnmsb.nxv8f16(<vscale x 8 x i1> [[TMP1]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
 ; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
 ;
   %1 = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> %p)
   %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %1, <vscale x 8 x half> %a, <vscale x 8 x half> %b)

 declare <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
 declare <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
 declare <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
 
 attributes #0 = { "target-features"="+sve" }