Optimize loading of Relocation/Symbol tables

xoofx · Oct 15, 2024 · 0c6bcb3 · 0c6bcb3
1 parent dcfc2fb
commit 0c6bcb3
Show file tree

Hide file tree

Showing 9 changed files with 237 additions and 95 deletions.
diff --git a/src/LibObjectFile.Bench/LibObjectFile.Bench.csproj b/src/LibObjectFile.Bench/LibObjectFile.Bench.csproj
@@ -0,0 +1,14 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net8.0</TargetFramework>
+    <ImplicitUsings>enable</ImplicitUsings>
+    <Nullable>enable</Nullable>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <ProjectReference Include="..\LibObjectFile\LibObjectFile.csproj" />
+  </ItemGroup>
+
+</Project>
diff --git a/src/LibObjectFile.Bench/Program.cs b/src/LibObjectFile.Bench/Program.cs
@@ -0,0 +1,58 @@
+// Copyright (c) Alexandre Mutel. All rights reserved.
+// This file is licensed under the BSD-Clause 2 license.
+// See the license.txt file in the project root for more information.
+
+using System.Diagnostics;
+using LibObjectFile.Elf;
+
+namespace LibObjectFile.Bench;
+
+internal class Program
+{
+    static void Main(string[] args)
+    {
+        var clock = Stopwatch.StartNew();
+        //var memoryStream = new MemoryStream();
+        foreach (var file in GetLinuxBins())
+        {
+            //memoryStream.SetLength(0);
+            using var stream = File.OpenRead((string)file[0]);
+            //stream.CopyTo(memoryStream);
+
+            if (ElfFile.IsElf(stream))
+            {
+                ElfFile.Read(stream);
+            }
+        }
+        clock.Stop();
+        Console.WriteLine($"{clock.Elapsed.TotalMilliseconds}ms");
+    }
+
+    public static IEnumerable<object[]> GetLinuxBins()
+    {
+        var wslDirectory = @"\\wsl$\Ubuntu\usr\bin";
+        if (OperatingSystem.IsLinux())
+        {
+            foreach (var file in Directory.EnumerateFiles(@"/usr/bin"))
+            {
+                yield return new object[] { file };
+            }
+        }
+        else if (OperatingSystem.IsWindows() && Directory.Exists(wslDirectory))
+        {
+            foreach (var file in Directory.EnumerateFiles(wslDirectory))
+            {
+                var fileInfo = new FileInfo(file);
+                // Skip symbolic links as loading them will fail
+                if ((fileInfo.Attributes & FileAttributes.ReparsePoint) == 0)
+                {
+                    yield return new object[] { file };
+                }
+            }
+        }
+        else
+        {
+            yield return new object[] { string.Empty };
+        }
+    }
+}
diff --git a/src/LibObjectFile.sln b/src/LibObjectFile.sln
@@ -30,6 +30,8 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "doc", "doc", "{BD580DD4-4E2
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "objdasm", "objdasm\objdasm.csproj", "{056AA737-6B5F-47A6-8426-E7918D930C5C}"
 EndProject
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "LibObjectFile.Bench", "LibObjectFile.Bench\LibObjectFile.Bench.csproj", "{34AD50B2-FAE3-42C9-8117-B5DE1CEEF0EA}"
+EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution
 		Debug|Any CPU = Debug|Any CPU
@@ -52,6 +54,10 @@ Global
 		{056AA737-6B5F-47A6-8426-E7918D930C5C}.Debug|Any CPU.Build.0 = Debug|Any CPU
 		{056AA737-6B5F-47A6-8426-E7918D930C5C}.Release|Any CPU.ActiveCfg = Release|Any CPU
 		{056AA737-6B5F-47A6-8426-E7918D930C5C}.Release|Any CPU.Build.0 = Release|Any CPU
+		{34AD50B2-FAE3-42C9-8117-B5DE1CEEF0EA}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{34AD50B2-FAE3-42C9-8117-B5DE1CEEF0EA}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{34AD50B2-FAE3-42C9-8117-B5DE1CEEF0EA}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{34AD50B2-FAE3-42C9-8117-B5DE1CEEF0EA}.Release|Any CPU.Build.0 = Release|Any CPU
 	EndGlobalSection
 	GlobalSection(SolutionProperties) = preSolution
 		HideSolutionNode = FALSE

diff --git a/src/LibObjectFile/Elf/Sections/ElfRelocation.cs b/src/LibObjectFile/Elf/Sections/ElfRelocation.cs
@@ -8,7 +8,7 @@ namespace LibObjectFile.Elf;
 /// A relocation entry in the <see cref="ElfRelocationTable"/>
 /// This is the value seen in <see cref="ElfNative.Elf32_Rel"/> or <see cref="ElfNative.Elf64_Rel"/>
 /// </summary>
-public record ElfRelocation
+public record struct ElfRelocation
 {
     public ElfRelocation()
     {

diff --git a/src/LibObjectFile/Elf/Sections/ElfRelocationTable.cs b/src/LibObjectFile/Elf/Sections/ElfRelocationTable.cs
@@ -4,7 +4,10 @@
 
 using System;
 using System.Collections.Generic;
+using System.Runtime.CompilerServices;
+using System.Runtime.InteropServices;
 using LibObjectFile.Diagnostics;
+using LibObjectFile.IO;
 
 namespace LibObjectFile.Elf;
 
@@ -63,95 +66,77 @@ public override void Write(ElfWriter writer)
     private unsafe void Read32(ElfReader reader)
     {
         var numberOfEntries = base.Size / base.TableEntrySize;
-        _entries.Capacity = (int)numberOfEntries;
+        var entries = _entries;
+        CollectionsMarshal.SetCount(entries, (int)numberOfEntries);
+        var span = CollectionsMarshal.AsSpan(entries);
+
         if (IsRelocationWithAddends)
         {
-            for (ulong i = 0; i < numberOfEntries; i++)
+            using var batch = new BatchDataReader<ElfNative.Elf32_Rela>(reader.Stream, (int)numberOfEntries);
+            ref var entry = ref MemoryMarshal.GetReference(span);
+            while (batch.HasNext())
             {
-                ElfNative.Elf32_Rela rel;
-                ulong streamOffset = (ulong)reader.Stream.Position;
-                if (!reader.TryReadData(sizeof(ElfNative.Elf32_Rela), out rel))
-                {
-                    reader.Diagnostics.Error(DiagnosticId.ELF_ERR_IncompleteRelocationAddendsEntry32Size, $"Unable to read entirely the relocation entry [{i}] from {Type} section [{Index}]. Not enough data (size: {base.TableEntrySize}) read at offset {streamOffset} from the stream");
-                }
-
-                var offset = reader.Decode(rel.r_offset);
+                ref var rel = ref batch.ReadNext();
+                entry.Offset = reader.Decode(rel.r_offset);
                 var r_info = reader.Decode(rel.r_info);
-                var type = new ElfRelocationType(Parent!.Arch, r_info & 0xFF);
-                var symbolIndex = r_info >> 8;
-                var addend = reader.Decode(rel.r_addend);
-
-                var entry = new ElfRelocation(offset, type, symbolIndex, addend);
-                _entries.Add(entry);
+                entry.Type = new ElfRelocationType(Parent!.Arch, r_info & 0xFF);
+                entry.SymbolIndex = r_info >> 8;
+                entry.Addend = reader.Decode(rel.r_addend);
+                entry = ref Unsafe.Add(ref entry, 1);
             }
         }
         else
         {
-            for (ulong i = 0; i < numberOfEntries; i++)
+            using var batch = new BatchDataReader<ElfNative.Elf32_Rel>(reader.Stream, (int)numberOfEntries);
+            ref var entry = ref MemoryMarshal.GetReference(span);
+            while (batch.HasNext())
             {
-                ElfNative.Elf32_Rel rel;
-                ulong streamOffset = (ulong)reader.Stream.Position;
-                if (!reader.TryReadData(sizeof(ElfNative.Elf32_Rel), out rel))
-                {
-                    reader.Diagnostics.Error(DiagnosticId.ELF_ERR_IncompleteRelocationEntry32Size, $"Unable to read entirely the relocation entry [{i}] from {Type} section [{Index}]. Not enough data (size: {base.TableEntrySize}) read at offset {streamOffset} from the stream");
-                }
-
-                var offset = reader.Decode(rel.r_offset);
-
+                ref var rel = ref batch.ReadNext();
+                entry.Offset = reader.Decode(rel.r_offset);
                 var r_info = reader.Decode(rel.r_info);
-                var type = new ElfRelocationType(Parent!.Arch, r_info & 0xFF);
-                var symbolIndex = r_info >> 8;
-
-                var entry = new ElfRelocation(offset, type, symbolIndex, 0);
-                _entries.Add(entry);
+                entry.Type = new ElfRelocationType(Parent!.Arch, r_info & 0xFF);
+                entry.SymbolIndex = r_info >> 8;
+                entry.Addend = 0;
+                entry = ref Unsafe.Add(ref entry, 1);
             }
         }
     }
 
     private unsafe void Read64(ElfReader reader)
     {
         var numberOfEntries = base.Size / base.TableEntrySize;
+        var entries = _entries;
+        CollectionsMarshal.SetCount(entries, (int)numberOfEntries);
+        var span = CollectionsMarshal.AsSpan(entries);
+
         if (IsRelocationWithAddends)
         {
-            for (ulong i = 0; i < numberOfEntries; i++)
+            using var batch = new BatchDataReader<ElfNative.Elf64_Rela>(reader.Stream, (int)numberOfEntries);
+            ref var entry = ref MemoryMarshal.GetReference(span);
+            while (batch.HasNext())
             {
-                ElfNative.Elf64_Rela rel;
-                ulong streamOffset = (ulong)reader.Stream.Position;
-                if (!reader.TryReadData(sizeof(ElfNative.Elf64_Rela), out rel))
-                {
-                    reader.Diagnostics.Error(DiagnosticId.ELF_ERR_IncompleteRelocationAddendsEntry64Size, $"Unable to read entirely the relocation entry [{i}] from {Type} section [{Index}]. Not enough data (size: {base.TableEntrySize}) read at offset {streamOffset} from the stream");
-                }
-
-                var offset = reader.Decode(rel.r_offset);
-
+                ref var rel = ref batch.ReadNext();
+                entry.Offset = reader.Decode(rel.r_offset);
                 var r_info = reader.Decode(rel.r_info);
-                var type = new ElfRelocationType(Parent!.Arch, (uint)(r_info & 0xFFFFFFFF));
-                var symbolIndex = (uint)(r_info >> 32);
-                var addend = reader.Decode(rel.r_addend);
-
-                var entry = new ElfRelocation(offset, type, symbolIndex, addend);
-                _entries.Add(entry);
+                entry.Type = new ElfRelocationType(Parent!.Arch, (uint)(r_info & 0xFFFFFFFF));
+                entry.SymbolIndex = (uint)(r_info >> 32);
+                entry.Addend = reader.Decode(rel.r_addend);
+                entry = ref Unsafe.Add(ref entry, 1);
             }
         }
         else
         {
-            for (ulong i = 0; i < numberOfEntries; i++)
+            using var batch = new BatchDataReader<ElfNative.Elf64_Rel>(reader.Stream, (int)numberOfEntries);
+            ref var entry = ref MemoryMarshal.GetReference(span);
+            while (batch.HasNext())
             {
-                ElfNative.Elf64_Rel rel;
-                ulong streamOffset = (ulong)reader.Stream.Position;
-                if (!reader.TryReadData(sizeof(ElfNative.Elf64_Rel), out rel))
-                {
-                    reader.Diagnostics.Error(DiagnosticId.ELF_ERR_IncompleteRelocationEntry64Size, $"Unable to read entirely the relocation entry [{i}] from {Type} section [{Index}]. Not enough data (size: {base.TableEntrySize}) read at offset {streamOffset} from the stream");
-                }
-
-                var offset = reader.Decode(rel.r_offset);
-
+                ref var rel = ref batch.ReadNext();
+                entry.Offset = reader.Decode(rel.r_offset);
                 var r_info = reader.Decode(rel.r_info);
-                var type = new ElfRelocationType(Parent!.Arch, (uint)(r_info & 0xFFFFFFFF));
-                var symbolIndex = (uint)(r_info >> 32);
-
-                var entry = new ElfRelocation(offset, type, symbolIndex, 0);
-                _entries.Add(entry);
+                entry.Type = new ElfRelocationType(Parent!.Arch, (uint)(r_info & 0xFFFFFFFF));
+                entry.SymbolIndex = (uint)(r_info >> 32);
+                entry.Addend = 0;
+                entry = ref Unsafe.Add(ref entry, 1);
             }
         }
     }

diff --git a/src/LibObjectFile/Elf/Sections/ElfSymbol.cs b/src/LibObjectFile/Elf/Sections/ElfSymbol.cs
@@ -10,7 +10,7 @@ namespace LibObjectFile.Elf;
 /// A symbol entry in the <see cref="ElfSymbolTable"/>
 /// This is the value seen in <see cref="ElfNative.Elf32_Sym"/> or <see cref="ElfNative.Elf64_Sym"/>
 /// </summary>
-public record ElfSymbol
+public record struct ElfSymbol
 {
     /// <summary>
     /// Gets or sets the value associated to this symbol.

diff --git a/src/LibObjectFile/Elf/Sections/ElfSymbolTable.cs b/src/LibObjectFile/Elf/Sections/ElfSymbolTable.cs
@@ -4,7 +4,10 @@
 
 using System;
 using System.Collections.Generic;
+using System.Runtime.CompilerServices;
+using System.Runtime.InteropServices;
 using LibObjectFile.Diagnostics;
+using LibObjectFile.IO;
 
 namespace LibObjectFile.Elf;
 
@@ -36,13 +39,17 @@ public override void Read(ElfReader reader)
         reader.Position = Position;
         Entries.Clear();
 
+        var numberOfEntries = (int)(base.Size / base.TableEntrySize);
+        var entries = Entries;
+        CollectionsMarshal.SetCount(entries, numberOfEntries);
+
         if (_is32)
         {
-            Read32(reader);
+            Read32(reader, numberOfEntries);
         }
         else
         {
-            Read64(reader);
+            Read64(reader, numberOfEntries);
         }
     }
 
@@ -58,20 +65,15 @@ public override void Write(ElfWriter writer)
         }
     }
 
-    private void Read32(ElfReader reader)
+    private void Read32(ElfReader reader, int numberOfEntries)
     {
-        var numberOfEntries = base.Size / base.TableEntrySize;
-        Entries.Capacity = (int)numberOfEntries;
-        for (ulong i = 0; i < numberOfEntries; i++)
+        using var batch = new BatchDataReader<ElfNative.Elf32_Sym>(reader.Stream, numberOfEntries);
+        var span = CollectionsMarshal.AsSpan(Entries);
+        ref var entry = ref MemoryMarshal.GetReference(span);
+        while (batch.HasNext())
         {
-            ElfNative.Elf32_Sym sym;
-            ulong streamOffset = (ulong)reader.Stream.Position;
-            if (!reader.TryReadData((int)base.TableEntrySize, out sym))
-            {
-                reader.Diagnostics.Error(DiagnosticId.ELF_ERR_IncompleteSymbolEntry32Size, $"Unable to read entirely the symbol entry [{i}] from {Type} section [{Index}]. Not enough data (size: {base.TableEntrySize}) read at offset {streamOffset} from the stream");
-            }
+            ref var sym = ref batch.ReadNext();
 
-            var entry = new ElfSymbol();
             entry.Name = new ElfString(reader.Decode(sym.st_name));
             entry.Value = reader.Decode(sym.st_value);
             entry.Size = reader.Decode(sym.st_size);
@@ -81,25 +83,19 @@ private void Read32(ElfReader reader)
             entry.Bind = (ElfSymbolBind)(st_info >> 4);
             entry.Visibility = (ElfSymbolVisibility) sym.st_other;
             entry.SectionLink = new ElfSectionLink(reader.Decode(sym.st_shndx));
-
-            Entries.Add(entry);
+            entry = ref Unsafe.Add(ref entry, 1);
         }
     }
 
-    private void Read64(ElfReader reader)
+    private void Read64(ElfReader reader, int numberOfEntries)
     {
-        var numberOfEntries = base.Size / base.TableEntrySize;
-        Entries.Capacity = (int)numberOfEntries;
-        for (ulong i = 0; i < numberOfEntries; i++)
+        using var batch = new BatchDataReader<ElfNative.Elf64_Sym>(reader.Stream, numberOfEntries);
+        var span = CollectionsMarshal.AsSpan(Entries);
+        ref var entry = ref MemoryMarshal.GetReference(span);
+        while (batch.HasNext())
         {
-            ElfNative.Elf64_Sym sym;
-            ulong streamOffset = (ulong)reader.Stream.Position;
-            if (!reader.TryReadData((int)base.TableEntrySize, out sym))
-            {
-                reader.Diagnostics.Error(DiagnosticId.ELF_ERR_IncompleteSymbolEntry64Size, $"Unable to read entirely the symbol entry [{i}] from {Type} section [{Index}]. Not enough data (size: {base.TableEntrySize}) read at offset {streamOffset} from the stream");
-            }
+            ref var sym = ref batch.ReadNext();
 
-            var entry = new ElfSymbol();
             entry.Name = new ElfString(reader.Decode(sym.st_name));
             entry.Value = reader.Decode(sym.st_value);
             entry.Size = reader.Decode(sym.st_size);
@@ -109,12 +105,10 @@ private void Read64(ElfReader reader)
             entry.Bind = (ElfSymbolBind)(st_info >> 4);
             entry.Visibility = (ElfSymbolVisibility)sym.st_other;
             entry.SectionLink = new ElfSectionLink(reader.Decode(sym.st_shndx));
-
-            Entries.Add(entry);
+            entry = ref Unsafe.Add(ref entry, 1);
         }
     }
-
-
+
     private void Write32(ElfWriter writer)
     {
         var stringTable = (ElfStringTable)Link.Section!;