[ARM64/Windows] optimal value of CopyThreshold for Windows after performance analysis...

diff --git a/src/mscorlib/src/System/Buffer.cs b/src/mscorlib/src/System/Buffer.cs
index 1a5a4b6ad95c01266ec6ca925926f2cccc1c2c52..b21ee7132f4c93e68ade9e941bfaa53f3168ca22 100644 (file)
--- a/src/mscorlib/src/System/Buffer.cs
+++ b/src/mscorlib/src/System/Buffer.cs
@@ -263,9 +263,9 @@ namespace System
             const nuint CopyThreshold = 2048;
 #elif ARM64
 #if PLATFORM_WINDOWS
-            // TODO-ARM64-WINDOWS-OPT determine optimal value for Windows
+            // Determined optimal value for Windows.
             // https://github.com/dotnet/coreclr/issues/13843
-            const nuint CopyThreshold = 2048;
+            const nuint CopyThreshold = UInt64.MaxValue;
 #else // PLATFORM_WINDOWS
             // Managed code is currently faster than glibc unoptimized memmove
             // TODO-ARM64-UNIX-OPT revisit when glibc optimized memmove is in Linux distros