Record CPU and analyze CPU performance counters using ETW

This change adds a batch file that records CPU performance counters on every context switch, a python script that analyzes the results, and a test program that abuses the branch predictor to make interesting results. Blog post to follow.
google · Nov 24, 2016 · 3f57bef · Trass3r · Apr 3, 2017 · randomascii
1 parent 092fa4d
commit 3f57bef
Show file tree

Hide file tree

Showing 9 changed files with 462 additions and 0 deletions.
diff --git a/LabScripts/ETWPMCDemo/.gitignore b/LabScripts/ETWPMCDemo/.gitignore
@@ -0,0 +1 @@
+pmc_counters_test.txt
diff --git a/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.cpp b/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.cpp
@@ -0,0 +1,30 @@
+#include <stdlib.h>
+#include <string>
+#include <algorithm>
+
+int sum_array(unsigned char* p, size_t count);
+
+int main(int argc, char* argv[])
+{
+	int64_t start = __rdtsc();
+	unsigned char buffer[8192];
+
+	for (auto& x : buffer)
+	{
+		x = (unsigned char)((rand() / 71) & 255);
+	}
+
+	if (argc > 1 && strcmp(argv[1], "-sort") == 0)
+		std::sort(buffer, buffer + sizeof(buffer));
+
+	int64_t mid = __rdtsc();
+	int total = 0;
+	for (int i = 0; i < 30000; ++i)
+	{
+		total += sum_array(buffer, sizeof(buffer));
+	}
+	int64_t end = __rdtsc();
+	printf("%5.2f MCycles for initialization and %5.2f MCycles for conditional adding.\n", (mid - start) / 1e6, (end - mid) / 1e6);
+
+	return 0;
+}
diff --git a/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.sln b/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.sln
@@ -0,0 +1,28 @@
+
+Microsoft Visual Studio Solution File, Format Version 12.00
+# Visual Studio 14
+VisualStudioVersion = 14.0.25420.1
+MinimumVisualStudioVersion = 10.0.40219.1
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "ConditionalCount", "ConditionalCount.vcxproj", "{A208226C-6D79-4317-B31E-732B6F57BE45}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|x64 = Debug|x64
+		Debug|x86 = Debug|x86
+		Release|x64 = Release|x64
+		Release|x86 = Release|x86
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Debug|x64.ActiveCfg = Debug|x64
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Debug|x64.Build.0 = Debug|x64
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Debug|x86.ActiveCfg = Debug|Win32
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Debug|x86.Build.0 = Debug|Win32
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Release|x64.ActiveCfg = Release|x64
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Release|x64.Build.0 = Release|x64
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Release|x86.ActiveCfg = Release|Win32
+		{A208226C-6D79-4317-B31E-732B6F57BE45}.Release|x86.Build.0 = Release|Win32
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+EndGlobal
diff --git a/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.vcxproj b/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.vcxproj
@@ -0,0 +1,154 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project DefaultTargets="Build" ToolsVersion="14.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup Label="ProjectConfigurations">
+    <ProjectConfiguration Include="Debug|Win32">
+      <Configuration>Debug</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Release|Win32">
+      <Configuration>Release</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Debug|x64">
+      <Configuration>Debug</Configuration>
+      <Platform>x64</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Release|x64">
+      <Configuration>Release</Configuration>
+      <Platform>x64</Platform>
+    </ProjectConfiguration>
+  </ItemGroup>
+  <PropertyGroup Label="Globals">
+    <ProjectGuid>{A208226C-6D79-4317-B31E-732B6F57BE45}</ProjectGuid>
+    <Keyword>Win32Proj</Keyword>
+    <RootNamespace>ConditionalCount</RootNamespace>
+    <WindowsTargetPlatformVersion>8.1</WindowsTargetPlatformVersion>
+  </PropertyGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <PlatformToolset>v140</PlatformToolset>
+    <CharacterSet>Unicode</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>false</UseDebugLibraries>
+    <PlatformToolset>v140</PlatformToolset>
+    <WholeProgramOptimization>false</WholeProgramOptimization>
+    <CharacterSet>Unicode</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <PlatformToolset>v140</PlatformToolset>
+    <CharacterSet>Unicode</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>false</UseDebugLibraries>
+    <PlatformToolset>v140</PlatformToolset>
+    <WholeProgramOptimization>true</WholeProgramOptimization>
+    <CharacterSet>Unicode</CharacterSet>
+  </PropertyGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
+  <ImportGroup Label="ExtensionSettings">
+  </ImportGroup>
+  <ImportGroup Label="Shared">
+  </ImportGroup>
+  <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 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>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
+    <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|x64'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <PropertyGroup Label="UserMacros" />
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <LinkIncremental>true</LinkIncremental>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
+    <LinkIncremental>true</LinkIncremental>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <LinkIncremental>false</LinkIncremental>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
+    <LinkIncremental>false</LinkIncremental>
+  </PropertyGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <ClCompile>
+      <PrecompiledHeader>
+      </PrecompiledHeader>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+      <PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <SubSystem>Console</SubSystem>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+    </Link>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
+    <ClCompile>
+      <PrecompiledHeader>
+      </PrecompiledHeader>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+      <PreprocessorDefinitions>_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <SubSystem>Console</SubSystem>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+    </Link>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <PrecompiledHeader>
+      </PrecompiledHeader>
+      <Optimization>MaxSpeed</Optimization>
+      <FunctionLevelLinking>true</FunctionLevelLinking>
+      <IntrinsicFunctions>true</IntrinsicFunctions>
+      <PreprocessorDefinitions>WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <SubSystem>Console</SubSystem>
+      <EnableCOMDATFolding>true</EnableCOMDATFolding>
+      <OptimizeReferences>true</OptimizeReferences>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+    </Link>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <PrecompiledHeader>
+      </PrecompiledHeader>
+      <Optimization>MaxSpeed</Optimization>
+      <FunctionLevelLinking>true</FunctionLevelLinking>
+      <IntrinsicFunctions>true</IntrinsicFunctions>
+      <PreprocessorDefinitions>NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <SubSystem>Console</SubSystem>
+      <EnableCOMDATFolding>true</EnableCOMDATFolding>
+      <OptimizeReferences>true</OptimizeReferences>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+    </Link>
+  </ItemDefinitionGroup>
+  <ItemGroup>
+    <Text Include="ReadMe.txt" />
+  </ItemGroup>
+  <ItemGroup>
+    <ClCompile Include="ConditionalCount.cpp" />
+    <ClCompile Include="OtherFile.cpp" />
+  </ItemGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
+  <ImportGroup Label="ExtensionTargets">
+  </ImportGroup>
+</Project>
diff --git a/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.vcxproj.filters b/LabScripts/ETWPMCDemo/ConditionalCount/ConditionalCount.vcxproj.filters
@@ -0,0 +1,28 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup>
+    <Filter Include="Source Files">
+      <UniqueIdentifier>{4FC737F1-C7A5-4376-A066-2A32D752A2FF}</UniqueIdentifier>
+      <Extensions>cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx</Extensions>
+    </Filter>
+    <Filter Include="Header Files">
+      <UniqueIdentifier>{93995380-89BD-4b04-88EB-625FBE52EBFB}</UniqueIdentifier>
+      <Extensions>h;hh;hpp;hxx;hm;inl;inc;xsd</Extensions>
+    </Filter>
+    <Filter Include="Resource Files">
+      <UniqueIdentifier>{67DA6AB6-F800-4c08-8B7A-83BB121AAD01}</UniqueIdentifier>
+      <Extensions>rc;ico;cur;bmp;dlg;rc2;rct;bin;rgs;gif;jpg;jpeg;jpe;resx;tiff;tif;png;wav;mfcribbon-ms</Extensions>
+    </Filter>
+  </ItemGroup>
+  <ItemGroup>
+    <Text Include="ReadMe.txt" />
+  </ItemGroup>
+  <ItemGroup>
+    <ClCompile Include="ConditionalCount.cpp">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="OtherFile.cpp">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+  </ItemGroup>
+</Project>
diff --git a/LabScripts/ETWPMCDemo/ConditionalCount/OtherFile.cpp b/LabScripts/ETWPMCDemo/ConditionalCount/OtherFile.cpp
@@ -0,0 +1,40 @@
+/*
+This exists as a separate source file, and Link Time Code Generation is
+disabled, in order to hide information from the optimizer so that it won't
+realize that this function is pure and idempotent.
+*/
+
+//#define	UNWOUND
+
+int sum_array(unsigned char* p, size_t count)
+{
+	int result = 0;
+#ifdef UNWOUND
+	// I expected this to have a higher percentage of mispredicted
+	// branches because four out of five branches in the loop are
+	// unpredictable (the loop-end branch is very predictable).
+	// But this code somehow shows almost zero branch mispredicts,
+	// and performance that is equivalent to the perfectly predicted
+	// code. Very odd.
+	for (size_t i = 0; i < count; i += 4)
+	{
+		if (p[i + 0] < 128)
+			result += p[i + 0];
+		if (p[i + 1] < 128)
+			result += p[i + 1];
+		if (p[i + 2] < 128)
+			result += p[i + 2];
+		if (p[i + 3] < 128)
+			result += p[i + 3];
+	}
+#else
+for (size_t i = 0; i < count; i += 1)
+{
+	// This test relies on this being implemented using a conditional
+	// branch instruction.
+	if (p[i] < 128)
+		result += p[i];
+}
+#endif
+	return result;
+}
diff --git a/LabScripts/ETWPMCDemo/ConditionalCount/ReadMe.txt b/LabScripts/ETWPMCDemo/ConditionalCount/ReadMe.txt
@@ -0,0 +1,4 @@
+This demonstrates how pre-sorting of an array can dramatically affect the number
+of branch mispredicts. The exact behavior is highly dependent on your compiler
+and CPU but this can be a good test case for testing CPU performance counters
+such as branch mispredict counts.