Add new PE CV_CFL_LANG language code for ALIASOBJ and Rust (#530)

* Add Language Code for PE

* Add Objective C baseline

* Add new PE CV_CFL_LANG language code for ALIASOBJ and RUST

* Update Rust string

* re-baseline with latest Sarif SDK

* fix comments

* fix comments for md file

* add . to end of sentence in md file

* FunctionalTestBuildScripts.md add spacing before code

* update FunctionalTestBuildScripts.md

* update to Rust 1.5.8.1

* Fix merge issue for release history file

* remove extra baseline expected sarif file

* Update ReleaseHistory.md

Co-authored-by: Eddy Nakamura <[email protected]>
Co-authored-by: Michael C. Fanning <[email protected]>

docs/FunctionalTestBuildScripts.md

@@ -0,0 +1,15 @@
+# Build Scripts
+
+This file records scripts used to compile the test files, in alphabetical order.
+Base scenario is a simple hello world program built with different parameters for testing purpose.
+Test files are located in [BaselineTestsData](https://github.com/microsoft/binskim/tree/main/src/Test.FunctionalTests.BinSkim.Driver/BaselineTestsData) and [FunctionalTestsData](https://github.com/microsoft/binskim/tree/main/src/Test.FunctionalTests.BinSkim.Rules/FunctionalTestsData).
+
+## clangcl.pe.c.codeview.exe
+
+A simple hello world program, compiled with `clang 13.0.0` that generates a .exe and associated .pdb file. Script to reproduce:  
+`clang-cl -o clangcl.pe.c.codeview.exe -fuse-ld=lld-link hello.c -m32 -Z7 -MTd`
+
+## Native_x64_RustC_Rust_debuginfo2_v1.58.1.exe
+
+A simple hello world program, compiled with `rustc 1.58.1` that generates a .exe and associated .pdb file. Script to reproduce:  
+`rustc -g -Clink-arg=/DEBUG:FULL src\main.rs -o Native_x64_RustC_Rust_debuginfo2_v1.58.1.exe`

src/BinaryParsers/PEBinary/ProgramDatabase/ObjectModuleDetails.cs

@@ -224,6 +224,11 @@ public string GetDialect(out string versionNumber)
         }
     }
 
+    /// <summary>
+    /// The CV_CFL_LANG enumeration, 
+    /// which specifies the code language of the application or linked module in the debug interface access SDK.
+    /// https://docs.microsoft.com/en-us/visualstudio/debugger/debug-interface-access/cv-cfl-lang
+    /// </summary>
     public enum Language : uint
     {
         C = 0x00,
@@ -242,11 +247,13 @@ public enum Language : uint
         Java = 0x0D,
         JScript = 0x0E,
         MSIL = 0x0F,
-        HLSL = 0x10, // High Level Shader Language
+        HLSL = 0x10,
         ObjectiveC = 0x11,
         ObjectiveCxx = 0x12,
         Swift = 0x13,
-        NASM = 0x4E, // The Netwide Assembler
+        ALIASOBJ = 0x14,
+        Rust = 0x15,
+        NASM = 0x4E,
         Unknown
     }
 

src/ReleaseHistory.md

@@ -1,5 +1,9 @@
 # BinSkim Release History
 
+## Unreleased
+
+* FEATURE: Add new PE `CV_CFL_LANG` language code for `ALIASOBJ` and `Rust`. [530](https://github.com/microsoft/binskim/pull/530)
+
 ## **v1.9.3** [NuGet Package](https://www.nuget.org/packages/Microsoft.CodeAnalysis.BinSkim/1.9.3)
 
 * BUGFIX: Fix `KeyNotFoundException` exception raised by `BA2006.BuildWithSecureTools` when individual `MinimumToolVersions` properties are removed from XML configuration. [#565](https://github.com/microsoft/binskim/pull/565)

src/Test.UnitTests.BinaryParsers/PEBinary/PEBinaryTests.cs

@@ -2,13 +2,17 @@
 // Licensed under the MIT license. See LICENSE file in the project root for full license information.
 
 using System;
+using System.Collections.Generic;
 using System.IO;
 using System.Reflection;
 
 using Dia2Lib;
 
 using FluentAssertions;
 
+using Microsoft.CodeAnalysis.BinaryParsers.ProgramDatabase;
+using Microsoft.CodeAnalysis.Sarif.Driver;
+
 using Xunit;
 
 namespace Microsoft.CodeAnalysis.BinaryParsers
@@ -70,6 +74,19 @@ public void PEBinary_PdbIsStripped()
             }
         }
 
+        [Fact]
+        public void PEBinary_ContainsExpectedLanguageCode()
+        {
+            if (!PlatformSpecificHelpers.RunningOnWindows()) { return; }
+
+            ContainsLanguageCode("clangcl.pe.c.codeview.exe", Language.C).Should().BeTrue();
+            ContainsLanguageCode("clangcl.pe.cpp.codeview.exe", Language.Cxx).Should().BeTrue();
+            // As of v1.58.1 Rust official compiler RustC does not yet use the new CV_CFL_LANG code for Rust.
+            // We will have another test file when new version use it.
+            ContainsLanguageCode("Native_x64_RustC_Rust_debuginfo2_v1.58.1.exe", Language.Rust).Should().BeFalse();
+            ContainsLanguageCode("Native_x64_VS2019_CPlusPlus_DEBUG_DEFAULT.dll", Language.Cxx).Should().BeTrue();
+        }
+
         [Fact]
         public void PEBinary_CanCreateIDiaSourceFromMsdia()
         {
@@ -92,5 +109,27 @@ public void PEBinary_TryLoadPdbFromSymbolFolder()
                 peBinary.PdbParseException.Should().BeNull();
             }
         }
+
+        private static bool ContainsLanguageCode(string fileName, Language language)
+        {
+            string fileFullPath = Path.Combine(TestData, "PE", fileName);
+            using (var peBinary = new PEBinary(new Uri(fileFullPath)))
+            {
+                peBinary.Valid.Should().BeTrue();
+                peBinary.Pdb.Should().NotBeNull();
+                peBinary.PdbParseException.Should().BeNull();
+
+                var languages = new HashSet<Language>();
+                foreach (DisposableEnumerableView<Symbol> omView in peBinary.Pdb.CreateObjectModuleIterator())
+                {
+                    ObjectModuleDetails omDetails = omView.Value.GetObjectModuleDetails();
+                    if (omDetails.Library == omDetails.Name)
+                    {
+                        languages.Add(omDetails.Language);
+                    }
+                }
+                return languages.Contains(language);
+            }
+        }
     }
 }