Optimize

ScriptableRenderPipeline/Core/CoreRP/ShaderLibrary/SpaceFillingCurves.hlsl

     return x;
 }
 
-// "Insert" two 0 bits after each of the 10 low bits of x/
+// "Insert" two 0 bits after each of the 10 low bits of x.
 // Ref: https://fgiesen.wordpress.com/2009/12/13/decoding-morton-codes/
 uint Part1By2(uint x)
 {
     return x;
 }
 
-// Inverse of Part1By1 - "delete" all odd-indexed bits/
+// Inverse of Part1By1 - "delete" all odd-indexed bits.
 // Ref: https://fgiesen.wordpress.com/2009/12/13/decoding-morton-codes/
 uint Compact1By1(uint x)
 {
     return x;
 }
 
-// Inverse of Part1By2 - "delete" all bits not at positions divisible by 3/
+// Inverse of Part1By2 - "delete" all bits not at positions divisible by 3.
 // Ref: https://fgiesen.wordpress.com/2009/12/13/decoding-morton-codes/
 uint Compact1By2(uint x)
 {

ScriptableRenderPipeline/HDRenderPipeline/HDRP/Material/SubsurfaceScattering/SubsurfaceScattering.compute

 void SubsurfaceScattering(uint2 groupId       : SV_GroupID,
                           uint  groupThreadId : SV_GroupThreadID)
 {
+    groupThreadId &= GROUP_SIZE_2D - 1; // Help the compiler
+
     // Note: any factor of 64 is a suitable wave size for our algorithm.
     uint waveIndex = WaveReadFirstLane(groupThreadId / 64);
     uint laneIndex = groupThreadId % 64;