Rewrite how Matrix3x2 and Matrix4x4 are implemented (#80091)

* Rewrite how Matrix3x2 and Matrix4x4 are implemented

* Fix a bug in lowerxarch related to merging Sse41.Insert chains

Author: tannergooding

src/coreclr/jit/lowerxarch.cpp

@@ -2809,6 +2809,9 @@ GenTree* Lowering::LowerHWIntrinsicCreate(GenTreeHWIntrinsic* node)
 
             if (comp->compOpportunisticallyDependsOn(InstructionSet_SSE41))
             {
+                assert(argCnt <= 4);
+                GenTree* insertedNodes[4];
+
                 for (N = 1; N < argCnt - 1; N++)
                 {
                     // We will be constructing the following parts:
@@ -2837,10 +2840,12 @@ GenTree* Lowering::LowerHWIntrinsicCreate(GenTreeHWIntrinsic* node)
                     idx = comp->gtNewIconNode(N << 4, TYP_INT);
                     BlockRange().InsertAfter(tmp2, idx);
 
-                    tmp1 = comp->gtNewSimdHWIntrinsicNode(simdType, tmp1, tmp2, idx, NI_SSE41_Insert, simdBaseJitType,
+                    tmp3 = comp->gtNewSimdHWIntrinsicNode(simdType, tmp1, tmp2, idx, NI_SSE41_Insert, simdBaseJitType,
                                                           simdSize);
-                    BlockRange().InsertAfter(idx, tmp1);
-                    LowerNode(tmp1);
+                    BlockRange().InsertAfter(idx, tmp3);
+
+                    insertedNodes[N] = tmp3;
+                    tmp1             = tmp3;
                 }
 
                 // We will be constructing the following parts:
@@ -2868,6 +2873,18 @@ GenTree* Lowering::LowerHWIntrinsicCreate(GenTreeHWIntrinsic* node)
                 BlockRange().InsertAfter(tmp2, idx);
 
                 node->ResetHWIntrinsicId(NI_SSE41_Insert, comp, tmp1, tmp2, idx);
+
+                for (N = 1; N < argCnt - 1; N++)
+                {
+                    // LowerNode for NI_SSE41_Insert specially handles zeros, constants, and certain mask values
+                    // to do the minimal number of operations and may merge together two neighboring inserts that
+                    // don't have any side effects between them. Because of this and because of the interdependence
+                    // of the inserts we've created above, we need to wait to lower the generated inserts until after
+                    // we've completed the chain.
+
+                    GenTree* insertedNode = insertedNodes[N];
+                    LowerNode(insertedNode);
+                }
                 break;
             }
 

src/libraries/System.Numerics.Vectors/tests/Matrix3x2Tests.cs

@@ -568,12 +568,15 @@ public void Matrix3x2EqualsTest()
         public void Matrix3x2GetHashCodeTest()
         {
             Matrix3x2 target = GenerateIncrementalMatrixNumber();
-            int expected = HashCode.Combine(target.M11, target.M12,
-                                     target.M21, target.M22,
-                                     target.M31, target.M32);
-            int actual;
 
-            actual = target.GetHashCode();
+            int expected = HashCode.Combine(
+                new Vector2(target.M11, target.M12),
+                new Vector2(target.M21, target.M22),
+                new Vector2(target.M31, target.M32)
+            );
+
+            int actual = target.GetHashCode();
+
             Assert.Equal(expected, actual);
         }
 

src/libraries/System.Numerics.Vectors/tests/Matrix4x4Tests.cs

@@ -1690,29 +1690,13 @@ public void Matrix4x4GetHashCodeTest()
         {
             Matrix4x4 target = GenerateIncrementalMatrixNumber();
 
-            HashCode hash = default;
+            int expected = HashCode.Combine(
+                new Vector4(target.M11, target.M12, target.M13, target.M14),
+                new Vector4(target.M21, target.M22, target.M23, target.M24),
+                new Vector4(target.M31, target.M32, target.M33, target.M34),
+                new Vector4(target.M41, target.M42, target.M43, target.M44)
+            );
 
-            hash.Add(target.M11);
-            hash.Add(target.M12);
-            hash.Add(target.M13);
-            hash.Add(target.M14);
-
-            hash.Add(target.M21);
-            hash.Add(target.M22);
-            hash.Add(target.M23);
-            hash.Add(target.M24);
-
-            hash.Add(target.M31);
-            hash.Add(target.M32);
-            hash.Add(target.M33);
-            hash.Add(target.M34);
-
-            hash.Add(target.M41);
-            hash.Add(target.M42);
-            hash.Add(target.M43);
-            hash.Add(target.M44);
-
-            int expected = hash.ToHashCode();
             int actual = target.GetHashCode();
 
             Assert.Equal(expected, actual);

src/libraries/System.Private.CoreLib/src/System.Private.CoreLib.Shared.projitems

@@ -556,7 +556,9 @@
     <Compile Include="$(MSBuildThisFileDirectory)System\Number.Parsing.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\BitOperations.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\Matrix3x2.cs" />
+    <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\Matrix3x2.Impl.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\Matrix4x4.cs" />
+    <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\Matrix4x4.Impl.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\Plane.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\Quaternion.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\TotalOrderIeee754Comparer.cs" />
@@ -2535,4 +2537,4 @@
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\IUnaryPlusOperators.cs" />
     <Compile Include="$(MSBuildThisFileDirectory)System\Numerics\IUnsignedNumber.cs" />
   </ItemGroup>
-</Project>
+</Project>