Windows profile build target (CPU profiling), and a little bit of optimization revealed by such.

4 files changed, 121 insertions, 8 deletions

diff --git a/node/Poly1305.cpp b/node/Poly1305.cpp
index b78071f6..13d4712d 100644
--- a/node/Poly1305.cpp
+++ b/node/Poly1305.cpp
@@ -135,11 +135,12 @@ typedef struct poly1305_context {
   unsigned char opaque[136];
 } poly1305_context;
 
-#if (defined(_MSC_VER) || defined(__GNUC__)) && (defined(__amd64) || defined(__amd64__) || defined(__x86_64) || defined(__x86_64__) || defined(__AMD64) || defined(__AMD64__))
+#if (defined(_MSC_VER) || defined(__GNUC__)) && (defined(__amd64) || defined(__amd64__) || defined(__x86_64) || defined(__x86_64__) || defined(__AMD64) || defined(__AMD64__) || defined(_M_X64))
 
 //////////////////////////////////////////////////////////////////////////////
 // 128-bit implementation for MSC and GCC from Poly1305-donna
 
+
 #if defined(_MSC_VER)
   #include <intrin.h>
 
@@ -183,9 +184,9 @@ typedef struct poly1305_state_internal_t {
   unsigned char final;
 } poly1305_state_internal_t;
 
-/* interpret eight 8 bit unsigned integers as a 64 bit unsigned integer in little endian */
-static inline unsigned long long
-U8TO64(const unsigned char *p) {
+#if defined(ZT_NO_TYPE_PUNNING) || (__BYTE_ORDER != __LITTLE_ENDIAN)
+static inline unsigned long long U8TO64(const unsigned char *p)
+{
   return
     (((unsigned long long)(p[0] & 0xff)      ) |
      ((unsigned long long)(p[1] & 0xff) <<  8) |
@@ -196,10 +197,13 @@ U8TO64(const unsigned char *p) {
      ((unsigned long long)(p[6] & 0xff) << 48) |
      ((unsigned long long)(p[7] & 0xff) << 56));
 }
+#else
+#define U8TO64(p) (*reinterpret_cast<const unsigned long long *>(p))
+#endif
 
-/* store a 64 bit unsigned integer as eight 8 bit unsigned integers in little endian */
-static inline void
-U64TO8(unsigned char *p, unsigned long long v) {
+#if defined(ZT_NO_TYPE_PUNNING) || (__BYTE_ORDER != __LITTLE_ENDIAN)
+static inline void U64TO8(unsigned char *p, unsigned long long v)
+{
   p[0] = (v      ) & 0xff;
   p[1] = (v >>  8) & 0xff;
   p[2] = (v >> 16) & 0xff;
@@ -209,6 +213,9 @@ U64TO8(unsigned char *p, unsigned long long v) {
   p[6] = (v >> 48) & 0xff;
   p[7] = (v >> 56) & 0xff;
 }
+#else
+#define U64TO8(p,v) ((*reinterpret_cast<unsigned long long *>(p)) = (v))
+#endif
 
 static inline void
 poly1305_init(poly1305_context *ctx, const unsigned char key[32]) {
diff --git a/windows/ZeroTierOne.sln b/windows/ZeroTierOne.sln
index 9baadd81..05f498fb 100644
--- a/windows/ZeroTierOne.sln
+++ b/windows/ZeroTierOne.sln
@@ -28,6 +28,10 @@ Global
 		DVD-5|Mixed Platforms = DVD-5|Mixed Platforms
 		DVD-5|Win32 = DVD-5|Win32
 		DVD-5|x64 = DVD-5|x64
+		Profile|Any CPU = Profile|Any CPU
+		Profile|Mixed Platforms = Profile|Mixed Platforms
+		Profile|Win32 = Profile|Win32
+		Profile|x64 = Profile|x64
 		Release|Any CPU = Release|Any CPU
 		Release|Mixed Platforms = Release|Mixed Platforms
 		Release|Win32 = Release|Win32
@@ -91,6 +95,13 @@ Global
 		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.DVD-5|x64.ActiveCfg = Debug|x64
 		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.DVD-5|x64.Build.0 = Debug|x64
 		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.DVD-5|x64.Deploy.0 = Debug|x64
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|Any CPU.ActiveCfg = Profile|Win32
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|Mixed Platforms.ActiveCfg = Profile|Win32
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|Mixed Platforms.Build.0 = Profile|Win32
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|Win32.ActiveCfg = Profile|Win32
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|Win32.Build.0 = Profile|Win32
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|x64.ActiveCfg = Profile|x64
+		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Profile|x64.Build.0 = Profile|x64
 		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Release|Any CPU.ActiveCfg = Release|Win32
 		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Release|Mixed Platforms.ActiveCfg = Release|Win32
 		{B00A4957-5977-4AC1-9EF4-571DC27EADA2}.Release|Mixed Platforms.Build.0 = Release|Win32
@@ -189,6 +200,10 @@ Global
 		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.DVD-5|x64.ActiveCfg = Win8 Release|x64
 		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.DVD-5|x64.Build.0 = Win8 Release|x64
 		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.DVD-5|x64.Deploy.0 = Win8 Release|x64
+		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Profile|Any CPU.ActiveCfg = Win8 Debug|x64
+		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Profile|Mixed Platforms.ActiveCfg = Win8 Debug|Win32
+		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Profile|Win32.ActiveCfg = Win8 Debug|Win32
+		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Profile|x64.ActiveCfg = Win8 Debug|x64
 		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Release|Any CPU.ActiveCfg = Win8 Release|Win32
 		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Release|Mixed Platforms.ActiveCfg = Win8 Release|Win32
 		{43BA7584-D4DB-4F7C-90FC-E2B18A68A213}.Release|Mixed Platforms.Build.0 = Win8 Release|Win32
@@ -287,6 +302,14 @@ Global
 		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.DVD-5|Mixed Platforms.Build.0 = Debug|Any CPU
 		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.DVD-5|Win32.ActiveCfg = Debug|Any CPU
 		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.DVD-5|x64.ActiveCfg = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|Any CPU.ActiveCfg = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|Any CPU.Build.0 = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|Mixed Platforms.ActiveCfg = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|Mixed Platforms.Build.0 = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|Win32.ActiveCfg = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|Win32.Build.0 = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|x64.ActiveCfg = Debug|Any CPU
+		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Profile|x64.Build.0 = Debug|Any CPU
 		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Release|Any CPU.ActiveCfg = Release|Any CPU
 		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Release|Any CPU.Build.0 = Release|Any CPU
 		{4CCA6B98-5E64-45BF-AC34-19B3E2570DB1}.Release|Mixed Platforms.ActiveCfg = Release|Any CPU
@@ -361,6 +384,14 @@ Global
 		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.DVD-5|Win32.Build.0 = Debug|Any CPU
 		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.DVD-5|x64.ActiveCfg = Debug|Any CPU
 		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.DVD-5|x64.Build.0 = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|Any CPU.ActiveCfg = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|Any CPU.Build.0 = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|Mixed Platforms.ActiveCfg = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|Mixed Platforms.Build.0 = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|Win32.ActiveCfg = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|Win32.Build.0 = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|x64.ActiveCfg = Debug|Any CPU
+		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Profile|x64.Build.0 = Debug|Any CPU
 		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Release|Any CPU.ActiveCfg = Release|Any CPU
 		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Release|Any CPU.Build.0 = Release|Any CPU
 		{6D27214A-087B-4484-B898-AD2A13FA3B9E}.Release|Mixed Platforms.ActiveCfg = Release|Any CPU
diff --git a/windows/ZeroTierOne/ZeroTierOne.vcxproj b/windows/ZeroTierOne/ZeroTierOne.vcxproj
index 32d7fa4a..84a44198 100644
--- a/windows/ZeroTierOne/ZeroTierOne.vcxproj
+++ b/windows/ZeroTierOne/ZeroTierOne.vcxproj
@@ -9,6 +9,14 @@
       <Configuration>Debug</Configuration>
       <Platform>x64</Platform>
     </ProjectConfiguration>
+    <ProjectConfiguration Include="Profile|Win32">
+      <Configuration>Profile</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Profile|x64">
+      <Configuration>Profile</Configuration>
+      <Platform>x64</Platform>
+    </ProjectConfiguration>
     <ProjectConfiguration Include="Release|Win32">
       <Configuration>Release</Configuration>
       <Platform>Win32</Platform>
@@ -66,6 +74,7 @@
     <ClCompile Include="..\..\node\Utils.cpp" />
     <ClCompile Include="..\..\one.cpp">
       <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">false</ExcludedFromBuild>
+      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Profile|x64'">false</ExcludedFromBuild>
     </ClCompile>
     <ClCompile Include="..\..\osdep\Http.cpp" />
     <ClCompile Include="..\..\osdep\ManagedRoute.cpp" />
@@ -113,6 +122,7 @@
     <ClInclude Include="..\..\node\Cluster.hpp" />
     <ClInclude Include="..\..\node\CMWC4096.hpp" />
     <ClInclude Include="..\..\node\Constants.hpp" />
+    <ClInclude Include="..\..\node\Credential.hpp" />
     <ClInclude Include="..\..\node\DeferredPackets.hpp" />
     <ClInclude Include="..\..\node\Dictionary.hpp" />
     <ClInclude Include="..\..\node\Hashtable.hpp" />
@@ -172,12 +182,24 @@
     <PlatformToolset>v140</PlatformToolset>
     <CharacterSet>MultiByte</CharacterSet>
   </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Profile|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <PlatformToolset>v140</PlatformToolset>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration">
     <ConfigurationType>Application</ConfigurationType>
     <UseDebugLibraries>true</UseDebugLibraries>
     <PlatformToolset>v140</PlatformToolset>
     <CharacterSet>MultiByte</CharacterSet>
   </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Profile|x64'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <PlatformToolset>v140</PlatformToolset>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration">
     <ConfigurationType>Application</ConfigurationType>
     <UseDebugLibraries>false</UseDebugLibraries>
@@ -198,9 +220,15 @@
   <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
     <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
   </ImportGroup>
+  <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Profile|Win32'" Label="PropertySheets">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
   <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="PropertySheets">
     <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
   </ImportGroup>
+  <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Profile|x64'" Label="PropertySheets">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
   <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
     <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
   </ImportGroup>
@@ -213,6 +241,11 @@
     <OutDir>$(SolutionDir)\Build\$(Platform)\$(Configuration)\</OutDir>
     <TargetName>zerotier-one_x86</TargetName>
   </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Profile|Win32'">
+    <TargetExt>.exe</TargetExt>
+    <OutDir>$(SolutionDir)\Build\$(Platform)\$(Configuration)\</OutDir>
+    <TargetName>zerotier-one_x86</TargetName>
+  </PropertyGroup>
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
     <TargetExt>.exe</TargetExt>
     <OutDir>$(SolutionDir)\Build\$(Platform)\$(Configuration)\</OutDir>
@@ -223,6 +256,11 @@
     <OutDir>$(SolutionDir)\Build\$(Platform)\$(Configuration)\</OutDir>
     <TargetName>zerotier-one_x64</TargetName>
   </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Profile|x64'">
+    <TargetExt>.exe</TargetExt>
+    <OutDir>$(SolutionDir)\Build\$(Platform)\$(Configuration)\</OutDir>
+    <TargetName>zerotier-one_x64</TargetName>
+  </PropertyGroup>
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
     <TargetExt>.exe</TargetExt>
     <OutDir>$(SolutionDir)\Build\$(Platform)\$(Configuration)\</OutDir>
@@ -244,6 +282,22 @@
       <ImageHasSafeExceptionHandlers>false</ImageHasSafeExceptionHandlers>
     </Link>
   </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Profile|Win32'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+      <SDLCheck>true</SDLCheck>
+      <AdditionalIncludeDirectories>
+      </AdditionalIncludeDirectories>
+      <PreprocessorDefinitions>NOMINMAX;STATICLIB;WIN32;ZT_TRACE;ZT_USE_MINIUPNPC;MINIUPNP_STATICLIB;ZT_SOFTWARE_UPDATE_DEFAULT="disable";%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <DisableSpecificWarnings>4996</DisableSpecificWarnings>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <AdditionalDependencies>wsock32.lib;ws2_32.lib;Iphlpapi.lib;Rpcrt4.lib;%(AdditionalDependencies)</AdditionalDependencies>
+      <ImageHasSafeExceptionHandlers>false</ImageHasSafeExceptionHandlers>
+    </Link>
+  </ItemDefinitionGroup>
   <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
     <ClCompile>
       <WarningLevel>Level3</WarningLevel>
@@ -262,6 +316,24 @@
       <AdditionalOptions>"notelemetry.obj" %(AdditionalOptions)</AdditionalOptions>
     </Link>
   </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Profile|x64'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+      <SDLCheck>true</SDLCheck>
+      <AdditionalIncludeDirectories>
+      </AdditionalIncludeDirectories>
+      <PreprocessorDefinitions>NOMINMAX;STATICLIB;WIN32;ZT_USE_MINIUPNPC;MINIUPNP_STATICLIB;ZT_SOFTWARE_UPDATE_DEFAULT="disable";%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <MultiProcessorCompilation>false</MultiProcessorCompilation>
+      <DisableSpecificWarnings>4996</DisableSpecificWarnings>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <AdditionalDependencies>wsock32.lib;ws2_32.lib;Iphlpapi.lib;Rpcrt4.lib;%(AdditionalDependencies)</AdditionalDependencies>
+      <ImageHasSafeExceptionHandlers>false</ImageHasSafeExceptionHandlers>
+      <AdditionalOptions>"notelemetry.obj" %(AdditionalOptions)</AdditionalOptions>
+    </Link>
+  </ItemDefinitionGroup>
   <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
     <ClCompile>
       <WarningLevel>Level3</WarningLevel>
@@ -291,7 +363,7 @@
   <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
     <ClCompile>
       <WarningLevel>Level3</WarningLevel>
-      <Optimization>MaxSpeed</Optimization>
+      <Optimization>Full</Optimization>
       <FunctionLevelLinking>true</FunctionLevelLinking>
       <IntrinsicFunctions>true</IntrinsicFunctions>
       <SDLCheck>true</SDLCheck>
diff --git a/windows/ZeroTierOne/ZeroTierOne.vcxproj.filters b/windows/ZeroTierOne/ZeroTierOne.vcxproj.filters
index ca1640e9..b051b4f7 100644
--- a/windows/ZeroTierOne/ZeroTierOne.vcxproj.filters
+++ b/windows/ZeroTierOne/ZeroTierOne.vcxproj.filters
@@ -482,6 +482,9 @@
     <ClInclude Include="..\..\node\CertificateOfOwnership.hpp">
       <Filter>Header Files\node</Filter>
     </ClInclude>
+    <ClInclude Include="..\..\node\Credential.hpp">
+      <Filter>Header Files\node</Filter>
+    </ClInclude>
   </ItemGroup>
   <ItemGroup>
     <ResourceCompile Include="ZeroTierOne.rc">