[Clang][LLVM][AArch64]SVE2.1 update the intrinsics according to acle[1] (#76844)

This patch changes the following intrinsic

 ```svst1uwq[_{d}]  replaced by svst1wq[_{d}]
 svst1uwq_vnum[_{d}] replaced by svst1wq_vnum[_{d}]
 svst1udq[_{d}]  replaced by svst1dq[_{d}]
 svst1udq_vnum[_{d}] replaced by svst1dq_vnum[_{d}]
```
Drops 'u' from the quadword stores because it is simply truncating the
quadwords to 32 bits

```
 svextq_lane[_{d}] replaced by  svextq[_{d}]
```
EXTQ follows the previous defined EXT intrinsics

```
 svdot[_{d}_{2}_{3}] replaced by svdot[_{d}_{2}]
```
Introduced with the latest SME2 ACLE change

[1]ARM-software/acle#257

Author: CarolineConcatto

clang/include/clang/Basic/arm_sve.td

@@ -454,11 +454,11 @@ let TargetGuard = "sve,bf16" in {
 
 let TargetGuard = "sve2p1" in {
   // Contiguous truncating store from quadword (single vector).
-  def SVST1UWQ      : MInst<"svst1uwq[_{d}]", "vPcd", "iUif",  [IsStore], MemEltTyInt32, "aarch64_sve_st1uwq">;
-  def SVST1UWQ_VNUM : MInst<"svst1uwq_vnum[_{d}]", "vPcld", "iUif", [IsStore], MemEltTyInt32, "aarch64_sve_st1uwq">;
+  def SVST1UWQ      : MInst<"svst1wq[_{d}]", "vPcd", "iUif",  [IsStore], MemEltTyInt32, "aarch64_sve_st1wq">;
+  def SVST1UWQ_VNUM : MInst<"svst1wq_vnum[_{d}]", "vPcld", "iUif", [IsStore], MemEltTyInt32, "aarch64_sve_st1wq">;
 
-  def SVST1UDQ      : MInst<"svst1udq[_{d}]", "vPcd", "lUld",  [IsStore], MemEltTyInt64, "aarch64_sve_st1udq">;
-  def SVST1UDQ_VNUM : MInst<"svst1udq_vnum[_{d}]", "vPcld", "lUld", [IsStore], MemEltTyInt64, "aarch64_sve_st1udq">;
+  def SVST1UDQ      : MInst<"svst1dq[_{d}]", "vPcd", "lUld",  [IsStore], MemEltTyInt64, "aarch64_sve_st1dq">;
+  def SVST1UDQ_VNUM : MInst<"svst1dq_vnum[_{d}]", "vPcld", "lUld", [IsStore], MemEltTyInt64, "aarch64_sve_st1dq">;
 
   // Store one vector (vector base + scalar offset)
   def SVST1Q_SCATTER_U64BASE_OFFSET : MInst<"svst1q_scatter[_{2}base]_offset[_{d}]",  "vPgld", "cUcsUsiUilUlfhdb", [IsScatterStore, IsByteIndexed], MemEltTyDefault, "aarch64_sve_st1q_scatter_scalar_offset">;
@@ -2040,12 +2040,12 @@ let TargetGuard = "sve2p1|sme2" in {
 }
 
 let TargetGuard = "sve2p1" in {
-def SVDOT_X2_S : SInst<"svdot[_{d}_{2}_{3}]", "ddhh", "i",  MergeNone, "aarch64_sve_sdot_x2", [], []>;
-def SVDOT_X2_U : SInst<"svdot[_{d}_{2}_{3}]", "ddhh", "Ui", MergeNone, "aarch64_sve_udot_x2", [], []>;
-def SVDOT_X2_F : SInst<"svdot[_{d}_{2}_{3}]", "ddhh", "f",  MergeNone, "aarch64_sve_fdot_x2", [], []>;
-def SVDOT_LANE_X2_S : SInst<"svdot_lane[_{d}_{2}_{3}]", "ddhhi", "i",  MergeNone, "aarch64_sve_sdot_lane_x2", [], [ImmCheck<3, ImmCheck0_3>]>;
-def SVDOT_LANE_X2_U : SInst<"svdot_lane[_{d}_{2}_{3}]", "ddhhi", "Ui", MergeNone, "aarch64_sve_udot_lane_x2", [], [ImmCheck<3, ImmCheck0_3>]>;
-def SVDOT_LANE_X2_F : SInst<"svdot_lane[_{d}_{2}_{3}]", "ddhhi", "f",  MergeNone, "aarch64_sve_fdot_lane_x2", [], [ImmCheck<3, ImmCheck0_3>]>;
+def SVDOT_X2_S : SInst<"svdot[_{d}_{2}]", "ddhh", "i",  MergeNone, "aarch64_sve_sdot_x2", [], []>;
+def SVDOT_X2_U : SInst<"svdot[_{d}_{2}]", "ddhh", "Ui", MergeNone, "aarch64_sve_udot_x2", [], []>;
+def SVDOT_X2_F : SInst<"svdot[_{d}_{2}]", "ddhh", "f",  MergeNone, "aarch64_sve_fdot_x2", [], []>;
+def SVDOT_LANE_X2_S : SInst<"svdot_lane[_{d}_{2}]", "ddhhi", "i",  MergeNone, "aarch64_sve_sdot_lane_x2", [], [ImmCheck<3, ImmCheck0_3>]>;
+def SVDOT_LANE_X2_U : SInst<"svdot_lane[_{d}_{2}]", "ddhhi", "Ui", MergeNone, "aarch64_sve_udot_lane_x2", [], [ImmCheck<3, ImmCheck0_3>]>;
+def SVDOT_LANE_X2_F : SInst<"svdot_lane[_{d}_{2}]", "ddhhi", "f",  MergeNone, "aarch64_sve_fdot_lane_x2", [], [ImmCheck<3, ImmCheck0_3>]>;
 }
 
 let TargetGuard = "sve2p1|sme2" in {
@@ -2208,7 +2208,7 @@ let TargetGuard = "sve2p1" in {
   def SVTBLQ : SInst<"svtblq[_{d}]", "ddu", "cUcsUsiUilUlbhfd", MergeNone, "aarch64_sve_tblq">;
   def SVTBXQ : SInst<"svtbxq[_{d}]", "dddu", "cUcsUsiUilUlbhfd", MergeNone, "aarch64_sve_tbxq">;
   // EXTQ
-  def EXTQ : SInst<"svextq_lane[_{d}]", "dddk", "cUcsUsiUilUlbhfd", MergeNone, "aarch64_sve_extq_lane", [], [ImmCheck<2, ImmCheck0_15>]>;
+  def EXTQ : SInst<"svextq[_{d}]", "dddk", "cUcsUsiUilUlbhfd", MergeNone, "aarch64_sve_extq", [], [ImmCheck<2, ImmCheck0_15>]>;
   // PMOV
   // Move to Pred
   multiclass PMOV_TO_PRED<string name, string types, string intrinsic, list<FlagType> flags=[], ImmCheckType immCh > {

clang/lib/CodeGen/CGBuiltin.cpp

@@ -9681,8 +9681,8 @@ Value *CodeGenFunction::EmitSVEMaskedStore(const CallExpr *E,
   bool IsQuadStore = false;
 
   switch (IntrinsicID) {
-  case Intrinsic::aarch64_sve_st1uwq:
-  case Intrinsic::aarch64_sve_st1udq:
+  case Intrinsic::aarch64_sve_st1wq:
+  case Intrinsic::aarch64_sve_st1dq:
     AddrMemoryTy = llvm::ScalableVectorType::get(MemEltTy, 1);
     PredTy =
         llvm::ScalableVectorType::get(IntegerType::get(getLLVMContext(), 1), 1);

clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_dot.c

@@ -26,7 +26,7 @@
 //
 svint32_t test_svdot_s32_x2(svint32_t op1, svint16_t op2, svint16_t op3)
 {
-  return SVE_ACLE_FUNC(svdot,_s32_s16_s16,)(op1, op2, op3);
+  return SVE_ACLE_FUNC(svdot,_s32_s16,)(op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svdot_u32_x2(
@@ -41,7 +41,7 @@ svint32_t test_svdot_s32_x2(svint32_t op1, svint16_t op2, svint16_t op3)
 //
 svuint32_t test_svdot_u32_x2(svuint32_t op1, svuint16_t op2, svuint16_t op3)
 {
-  return SVE_ACLE_FUNC(svdot,_u32_u16_u16,)(op1, op2, op3);
+  return SVE_ACLE_FUNC(svdot,_u32_u16,)(op1, op2, op3);
 }
 
 // CHECK-LABEL: @test_svdot_f32_x2(
@@ -56,7 +56,7 @@ svuint32_t test_svdot_u32_x2(svuint32_t op1, svuint16_t op2, svuint16_t op3)
 //
 svfloat32_t test_svdot_f32_x2(svfloat32_t op1, svfloat16_t op2, svfloat16_t op3)
 {
-  return SVE_ACLE_FUNC(svdot,_f32_f16_f16,)(op1, op2, op3);
+  return SVE_ACLE_FUNC(svdot,_f32_f16,)(op1, op2, op3);
 }
 
 
@@ -73,7 +73,7 @@ svfloat32_t test_svdot_f32_x2(svfloat32_t op1, svfloat16_t op2, svfloat16_t op3)
 //
 svint32_t test_svdot_lane_s32_x2(svint32_t op1, svint16_t op2, svint16_t op3)
 {
-  return SVE_ACLE_FUNC(svdot_lane,_s32_s16_s16,)(op1, op2, op3, 3);
+  return SVE_ACLE_FUNC(svdot_lane,_s32_s16,)(op1, op2, op3, 3);
 }
 
 // CHECK-LABEL: @test_svdot_lane_u32_x2(
@@ -88,7 +88,7 @@ svint32_t test_svdot_lane_s32_x2(svint32_t op1, svint16_t op2, svint16_t op3)
 //
 svuint32_t test_svdot_lane_u32_x2(svuint32_t op1, svuint16_t op2, svuint16_t op3)
 {
-  return SVE_ACLE_FUNC(svdot_lane,_u32_u16_u16,)(op1, op2, op3, 3);
+  return SVE_ACLE_FUNC(svdot_lane,_u32_u16,)(op1, op2, op3, 3);
 }
 
 // CHECK-LABEL: @test_svdot_lane_f32_x2(
@@ -103,5 +103,5 @@ svuint32_t test_svdot_lane_u32_x2(svuint32_t op1, svuint16_t op2, svuint16_t op3
 //
 svfloat32_t test_svdot_lane_f32_x2(svfloat32_t op1, svfloat16_t op2, svfloat16_t op3)
 {
-  return SVE_ACLE_FUNC(svdot_lane,_f32_f16_f16,)(op1, op2, op3, 3);
+  return SVE_ACLE_FUNC(svdot_lane,_f32_f16,)(op1, op2, op3, 3);
 }