Add support for Add Sve.GatherVectorUInt*ZeroExtendFirstFaulting()

src/coreclr/jit/gentree.cpp

@@ -26675,16 +26675,26 @@ bool GenTreeHWIntrinsic::OperIsMemoryLoad(GenTree** pAddr) const
 
             case NI_Sve_GatherVector:
             case NI_Sve_GatherVectorByteZeroExtend:
+            case NI_Sve_GatherVectorByteZeroExtendFirstFaulting:
             case NI_Sve_GatherVectorFirstFaulting:
             case NI_Sve_GatherVectorInt16SignExtend:
+            case NI_Sve_GatherVectorInt16SignExtendFirstFaulting:
             case NI_Sve_GatherVectorInt16WithByteOffsetsSignExtend:
+            case NI_Sve_GatherVectorInt16WithByteOffsetsSignExtendFirstFaulting:
             case NI_Sve_GatherVectorInt32SignExtend:
+            case NI_Sve_GatherVectorInt32SignExtendFirstFaulting:
             case NI_Sve_GatherVectorInt32WithByteOffsetsSignExtend:
+            case NI_Sve_GatherVectorInt32WithByteOffsetsSignExtendFirstFaulting:
             case NI_Sve_GatherVectorSByteSignExtend:
+            case NI_Sve_GatherVectorSByteSignExtendFirstFaulting:
             case NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtend:
+            case NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtendFirstFaulting:
             case NI_Sve_GatherVectorUInt16ZeroExtend:
+            case NI_Sve_GatherVectorUInt16ZeroExtendFirstFaulting:
             case NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtend:
+            case NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtendFirstFaulting:
             case NI_Sve_GatherVectorUInt32ZeroExtend:
+            case NI_Sve_GatherVectorUInt32ZeroExtendFirstFaulting:
             case NI_Sve_GatherVectorWithByteOffsetFirstFaulting:
             case NI_Sve_GatherVectorWithByteOffsets:
             case NI_Sve_LoadVector:
@@ -26806,14 +26816,24 @@ bool GenTreeHWIntrinsic::OperIsMemoryLoad(GenTree** pAddr) const
     {
 #ifdef TARGET_ARM64
         static_assert_no_msg(
-            AreContiguous(NI_Sve_GatherVector, NI_Sve_GatherVectorByteZeroExtend, NI_Sve_GatherVectorFirstFaulting,
-                          NI_Sve_GatherVectorInt16SignExtend, NI_Sve_GatherVectorInt16WithByteOffsetsSignExtend,
-                          NI_Sve_GatherVectorInt32SignExtend, NI_Sve_GatherVectorInt32WithByteOffsetsSignExtend,
-                          NI_Sve_GatherVectorSByteSignExtend, NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtend,
-                          NI_Sve_GatherVectorUInt16ZeroExtend, NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtend,
-                          NI_Sve_GatherVectorUInt32ZeroExtend));
-        assert(varTypeIsI(addr) || (varTypeIsSIMD(addr) && ((intrinsicId >= NI_Sve_GatherVector) &&
-                                                            (intrinsicId <= NI_Sve_GatherVectorUInt32ZeroExtend))));
+            AreContiguous(NI_Sve_GatherVector, NI_Sve_GatherVectorByteZeroExtend,
+                          NI_Sve_GatherVectorByteZeroExtendFirstFaulting, NI_Sve_GatherVectorFirstFaulting,
+                          NI_Sve_GatherVectorInt16SignExtend, NI_Sve_GatherVectorInt16SignExtendFirstFaulting,
+                          NI_Sve_GatherVectorInt16WithByteOffsetsSignExtend,
+                          NI_Sve_GatherVectorInt16WithByteOffsetsSignExtendFirstFaulting,
+                          NI_Sve_GatherVectorInt32SignExtend, NI_Sve_GatherVectorInt32SignExtendFirstFaulting,
+                          NI_Sve_GatherVectorInt32WithByteOffsetsSignExtend,
+                          NI_Sve_GatherVectorInt32WithByteOffsetsSignExtendFirstFaulting,
+                          NI_Sve_GatherVectorSByteSignExtend, NI_Sve_GatherVectorSByteSignExtendFirstFaulting,
+                          NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtend,
+                          NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtendFirstFaulting,
+                          NI_Sve_GatherVectorUInt16ZeroExtend, NI_Sve_GatherVectorUInt16ZeroExtendFirstFaulting,
+                          NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtend,
+                          NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtendFirstFaulting,
+                          NI_Sve_GatherVectorUInt32ZeroExtend, NI_Sve_GatherVectorUInt32ZeroExtendFirstFaulting));
+        assert(varTypeIsI(addr) ||
+               (varTypeIsSIMD(addr) && ((intrinsicId >= NI_Sve_GatherVector) &&
+                                        (intrinsicId <= NI_Sve_GatherVectorUInt32ZeroExtendFirstFaulting))));
 #else
         assert(varTypeIsI(addr));
 #endif

src/coreclr/jit/hwintrinsic.cpp

@@ -2175,18 +2175,28 @@ GenTree* Compiler::impHWIntrinsic(NamedIntrinsic        intrinsic,
 #elif defined(TARGET_ARM64)
                     case NI_Sve_GatherVector:
                     case NI_Sve_GatherVectorByteZeroExtend:
+                    case NI_Sve_GatherVectorByteZeroExtendFirstFaulting:
                     case NI_Sve_GatherVectorFirstFaulting:
                     case NI_Sve_GatherVectorInt16SignExtend:
+                    case NI_Sve_GatherVectorInt16SignExtendFirstFaulting:
                     case NI_Sve_GatherVectorInt16WithByteOffsetsSignExtend:
+                    case NI_Sve_GatherVectorInt16WithByteOffsetsSignExtendFirstFaulting:
                     case NI_Sve_GatherVectorInt32SignExtend:
+                    case NI_Sve_GatherVectorInt32SignExtendFirstFaulting:
                     case NI_Sve_GatherVectorInt32WithByteOffsetsSignExtend:
+                    case NI_Sve_GatherVectorInt32WithByteOffsetsSignExtendFirstFaulting:
                     case NI_Sve_GatherVectorSByteSignExtend:
+                    case NI_Sve_GatherVectorSByteSignExtendFirstFaulting:
                     case NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtend:
+                    case NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtendFirstFaulting:
                     case NI_Sve_GatherVectorUInt16ZeroExtend:
+                    case NI_Sve_GatherVectorUInt16ZeroExtendFirstFaulting:
                     case NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtend:
+                    case NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtendFirstFaulting:
                     case NI_Sve_GatherVectorUInt32ZeroExtend:
                     case NI_Sve_GatherVectorWithByteOffsetFirstFaulting:
                     case NI_Sve_GatherVectorWithByteOffsets:
+                    case NI_Sve_GatherVectorUInt32ZeroExtendFirstFaulting:
                         assert(varTypeIsSIMD(op3->TypeGet()));
                         if (numArgs == 3)
                         {

src/coreclr/jit/hwintrinsiccodegenarm64.cpp

@@ -2097,7 +2097,17 @@ void CodeGen::genHWIntrinsic(GenTreeHWIntrinsic* node)
                 break;
             }
 
+            case NI_Sve_GatherVectorByteZeroExtendFirstFaulting:
             case NI_Sve_GatherVectorFirstFaulting:
+            case NI_Sve_GatherVectorInt16SignExtendFirstFaulting:
+            case NI_Sve_GatherVectorInt16WithByteOffsetsSignExtendFirstFaulting:
+            case NI_Sve_GatherVectorInt32SignExtendFirstFaulting:
+            case NI_Sve_GatherVectorInt32WithByteOffsetsSignExtendFirstFaulting:
+            case NI_Sve_GatherVectorSByteSignExtendFirstFaulting:
+            case NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtendFirstFaulting:
+            case NI_Sve_GatherVectorUInt16ZeroExtendFirstFaulting:
+            case NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtendFirstFaulting:
+            case NI_Sve_GatherVectorUInt32ZeroExtendFirstFaulting:
             {
                 if (node->GetAuxiliaryType() == TYP_UNKNOWN)
                 {
@@ -2122,6 +2132,7 @@ void CodeGen::genHWIntrinsic(GenTreeHWIntrinsic* node)
                         GetEmitter()->emitIns_R(INS_sve_wrffr, emitSize, op4Reg, opt);
                     }
                 }
+
                 FALLTHROUGH;
             }
             case NI_Sve_GatherVector:
@@ -2144,7 +2155,9 @@ void CodeGen::genHWIntrinsic(GenTreeHWIntrinsic* node)
                     emitAttr baseSize = emitActualTypeSize(intrin.baseType);
                     bool     isLoadingBytes =
                         ((ins == INS_sve_ld1b) || (ins == INS_sve_ld1sb) || (ins == INS_sve_ldff1b) ||
-                         (ins == INS_sve_ldff1sb) || (intrin.id == NI_Sve_GatherVectorWithByteOffsetFirstFaulting));
+                         (ins == INS_sve_ldff1sb) || (intrin.id == NI_Sve_GatherVectorWithByteOffsetFirstFaulting) ||
+                         (intrin.id == NI_Sve_GatherVectorUInt32WithByteOffsetsZeroExtendFirstFaulting) ||
+                         (intrin.id == NI_Sve_GatherVectorUInt16WithByteOffsetsZeroExtendFirstFaulting));
                     insScalableOpts sopt = INS_SCALABLE_OPTS_NONE;
 
                     if (baseSize == EA_4BYTE)