[cdac] Implement GetObjectStringData (#105061)

elinor-fung · web-flow · commit 5fd965d7bf01 · 2024-07-19T06:25:47.000-07:00
- Include `String` and `Object` in data descriptor
- Start an `Object` contract for getting information about known managed objects
- Make cDAC implement `ISOSDacInterface::GetObjectStringData`
diff --git a/docs/design/datacontracts/Object.md b/docs/design/datacontracts/Object.md
@@ -0,0 +1,53 @@
+# Contract Object
+
+This contract is for getting information about well-known managed objects
+
+## APIs of contract
+
+``` csharp
+// Get the method table address for the object
+TargetPointer GetMethodTableAddress(TargetPointer address);
+
+// Get the string corresponding to a managed string object. Error if address does not represent a string.
+string GetStringValue(TargetPointer address);
+```
+
+## Version 1
+
+Data descriptors used:
+| Data Descriptor Name | Field | Meaning |
+| --- | --- | --- |
+| `Object` | `m_pMethTab` | Method table for the object |
+| `String` | `m_FirstChar` | First character of the string - `m_StringLength` can be used to read the full string (encoded in UTF-16) |
+| `String` | `m_StringLength` | Length of the string in characters (encoded in UTF-16) |
+
+Global variables used:
+| Global Name | Type | Purpose |
+| --- | --- | --- |
+| `ObjectToMethodTableUnmask` | uint8 | Bits to clear for converting to a method table address |
+| `StringMethodTable` | TargetPointer | The method table for System.String |
+
+``` csharp
+TargetPointer GetMethodTableAddress(TargetPointer address)
+{
+    TargetPointer mt = _targetPointer.ReadPointer(address + /* Object::m_pMethTab offset */);
+    return mt.Value & ~target.ReadGlobal<byte>("ObjectToMethodTableUnmask");
+}
+
+string GetStringValue(TargetPointer address)
+{
+    TargetPointer mt = GetMethodTableAddress(address);
+    TargetPointer stringMethodTable = target.ReadPointer(target.ReadGlobalPointer("StringMethodTable"));
+    if (mt != stringMethodTable)
+        throw new ArgumentException("Address does not represent a string object", nameof(address));
+
+    // Validates the method table
+    _ = target.Contracts.RuntimeTypeSystem.GetTypeHandle(mt);
+
+    Data.String str = _target.ProcessedData.GetOrAdd<Data.String>(address);
+    uint length = target.Read<uint>(address + /* String::m_StringLength offset */);
+    Span<byte> span = stackalloc byte[(int)length * sizeof(char)];
+    target.ReadBuffer(address + /* String::m_FirstChar offset */, span);
+    return new string(MemoryMarshal.Cast<byte, char>(span));
+}
+```
diff --git a/src/coreclr/debug/daccess/dacimpl.h b/src/coreclr/debug/daccess/dacimpl.h
@@ -1241,6 +1241,7 @@ class ClrDataAccess
     HRESULT GetMethodTableForEEClassImpl (CLRDATA_ADDRESS eeClassReallyMT, CLRDATA_ADDRESS *value);
     HRESULT GetMethodTableNameImpl(CLRDATA_ADDRESS mt, unsigned int count, _Inout_updates_z_(count) WCHAR *mtName, unsigned int *pNeeded);
     HRESULT GetObjectExceptionDataImpl(CLRDATA_ADDRESS objAddr, struct DacpExceptionObjectData *data);
+    HRESULT GetObjectStringDataImpl(CLRDATA_ADDRESS obj, unsigned int count, _Inout_updates_z_(count) WCHAR *stringData, unsigned int *pNeeded);
 
     BOOL IsExceptionFromManagedCode(EXCEPTION_RECORD * pExceptionRecord);
 #ifndef TARGET_UNIX
diff --git a/src/coreclr/debug/daccess/request.cpp b/src/coreclr/debug/daccess/request.cpp
@@ -1601,7 +1601,7 @@ ClrDataAccess::GetDomainFromContext(CLRDATA_ADDRESS contextAddr, CLRDATA_ADDRESS
 
 
 HRESULT
-ClrDataAccess::GetObjectStringData(CLRDATA_ADDRESS obj, unsigned int count, _Inout_updates_z_(count) WCHAR *stringData, unsigned int *pNeeded)
+ClrDataAccess::GetObjectStringData(CLRDATA_ADDRESS obj, unsigned int count, _Inout_updates_z_(count) WCHAR* stringData, unsigned int* pNeeded)
 {
     if (obj == 0)
         return E_INVALIDARG;
@@ -1611,44 +1611,73 @@ ClrDataAccess::GetObjectStringData(CLRDATA_ADDRESS obj, unsigned int count, _Ino
 
     SOSDacEnter();
 
+    if (m_cdacSos != NULL)
+    {
+        hr = m_cdacSos->GetObjectStringData(obj, count, stringData, pNeeded);
+        if (FAILED(hr))
+        {
+            hr = GetObjectStringDataImpl(obj, count, stringData, pNeeded);
+        }
+#ifdef _DEBUG
+        else
+        {
+            unsigned int neededLocal;
+            SString stringDataLocal;
+            HRESULT hrLocal = GetObjectStringDataImpl(obj, count, stringDataLocal.OpenUnicodeBuffer(count), &neededLocal);
+            _ASSERTE(hr == hrLocal);
+            _ASSERTE(pNeeded == NULL || *pNeeded == neededLocal);
+            _ASSERTE(u16_strncmp(stringData, stringDataLocal, count) == 0);
+        }
+#endif
+    }
+    else
+    {
+        hr = GetObjectStringDataImpl(obj, count, stringData, pNeeded);
+    }
+
+    SOSDacLeave();
+    return hr;
+}
+
+HRESULT
+ClrDataAccess::GetObjectStringDataImpl(CLRDATA_ADDRESS obj, unsigned int count, _Inout_updates_z_(count) WCHAR *stringData, unsigned int *pNeeded)
+{
     TADDR mtTADDR = DACGetMethodTableFromObjectPointer(TO_TADDR(obj), m_pTarget);
     PTR_MethodTable mt = PTR_MethodTable(mtTADDR);
 
     // Object must be a string
     BOOL bFree = FALSE;
     if (!DacValidateMethodTable(mt, bFree))
-        hr = E_INVALIDARG;
-    else if (HOST_CDADDR(mt) != HOST_CDADDR(g_pStringClass))
-        hr = E_INVALIDARG;
+        return E_INVALIDARG;
 
-    if (SUCCEEDED(hr))
-    {
-        PTR_StringObject str(TO_TADDR(obj));
-        ULONG32 needed = (ULONG32)str->GetStringLength() + 1;
+    if (HOST_CDADDR(mt) != HOST_CDADDR(g_pStringClass))
+        return E_INVALIDARG;
 
-        if (stringData && count > 0)
-        {
-            if (count > needed)
-                count = needed;
+    PTR_StringObject str(TO_TADDR(obj));
+    ULONG32 needed = (ULONG32)str->GetStringLength() + 1;
 
-            TADDR pszStr = TO_TADDR(obj)+offsetof(StringObject, m_FirstChar);
-            hr = m_pTarget->ReadVirtual(pszStr, (PBYTE)stringData, count * sizeof(WCHAR), &needed);
+    HRESULT hr;
+    if (stringData && count > 0)
+    {
+        if (count > needed)
+            count = needed;
 
-            if (SUCCEEDED(hr))
-                stringData[count - 1] = W('\0');
-            else
-                stringData[0] = W('\0');
-        }
-        else
-        {
-            hr = E_INVALIDARG;
-        }
+        TADDR pszStr = TO_TADDR(obj)+offsetof(StringObject, m_FirstChar);
+        hr = m_pTarget->ReadVirtual(pszStr, (PBYTE)stringData, count * sizeof(WCHAR), &needed);
 
-        if (pNeeded)
-            *pNeeded = needed;
+        if (SUCCEEDED(hr))
+            stringData[count - 1] = W('\0');
+        else
+            stringData[0] = W('\0');
+    }
+    else
+    {
+        hr = E_INVALIDARG;
     }
 
-    SOSDacLeave();
+    if (pNeeded)
+        *pNeeded = needed;
+
     return hr;
 }
 
diff --git a/src/coreclr/debug/runtimeinfo/contracts.jsonc b/src/coreclr/debug/runtimeinfo/contracts.jsonc
@@ -12,6 +12,7 @@
   "DacStreams": 1,
   "Exception": 1,
   "Loader": 1,
+  "Object": 1,
   "RuntimeTypeSystem": 1,
   "Thread": 1
 }
diff --git a/src/coreclr/debug/runtimeinfo/datadescriptor.h b/src/coreclr/debug/runtimeinfo/datadescriptor.h
@@ -172,6 +172,17 @@ CDAC_TYPE_BEGIN(GCHandle)
 CDAC_TYPE_SIZE(sizeof(OBJECTHANDLE))
 CDAC_TYPE_END(GCHandle)
 
+CDAC_TYPE_BEGIN(Object)
+CDAC_TYPE_INDETERMINATE(Object)
+CDAC_TYPE_FIELD(Object, /*pointer*/, m_pMethTab, cdac_offsets<Object>::m_pMethTab)
+CDAC_TYPE_END(Object)
+
+CDAC_TYPE_BEGIN(String)
+CDAC_TYPE_INDETERMINATE(String)
+CDAC_TYPE_FIELD(String, /*pointer*/, m_FirstChar, cdac_offsets<StringObject>::m_FirstChar)
+CDAC_TYPE_FIELD(String, /*uint32*/, m_StringLength, cdac_offsets<StringObject>::m_StringLength)
+CDAC_TYPE_END(String)
+
 // Loader
 
 CDAC_TYPE_BEGIN(Module)
@@ -264,8 +275,15 @@ CDAC_GLOBAL(FeatureEHFunclets, uint8, 1)
 #else
 CDAC_GLOBAL(FeatureEHFunclets, uint8, 0)
 #endif
+// See Object::GetGCSafeMethodTable
+#ifdef TARGET_64BIT
+CDAC_GLOBAL(ObjectToMethodTableUnmask, uint8, 1 | 1 << 1 | 1 << 2)
+#else
+CDAC_GLOBAL(ObjectToMethodTableUnmask, uint8, 1 | 1 << 1)
+#endif //TARGET_64BIT
 CDAC_GLOBAL(SOSBreakingChangeVersion, uint8, SOS_BREAKING_CHANGE_VERSION)
 CDAC_GLOBAL_POINTER(FreeObjectMethodTable, &::g_pFreeObjectMethodTable)
+CDAC_GLOBAL_POINTER(StringMethodTable, &::g_pStringClass)
 CDAC_GLOBAL_POINTER(MiniMetaDataBuffAddress, &::g_MiniMetaDataBuffAddress)
 CDAC_GLOBAL_POINTER(MiniMetaDataBuffMaxSize, &::g_MiniMetaDataBuffMaxSize)
 CDAC_GLOBALS_END()
diff --git a/src/coreclr/vm/object.h b/src/coreclr/vm/object.h
@@ -462,6 +462,14 @@ class Object
 
  private:
     VOID ValidateInner(BOOL bDeep, BOOL bVerifyNextHeader, BOOL bVerifySyncBlock);
+
+    template<typename T> friend struct ::cdac_offsets;
+};
+
+template<>
+struct cdac_offsets<Object>
+{
+    static constexpr size_t m_pMethTab = offsetof(Object, m_pMethTab);
 };
 
 /*
@@ -930,6 +938,15 @@ class StringObject : public Object
 private:
     static STRINGREF* EmptyStringRefPtr;
     static bool EmptyStringIsFrozen;
+
+    template<typename T> friend struct ::cdac_offsets;
+};
+
+template<>
+struct cdac_offsets<StringObject>
+{
+    static constexpr size_t m_FirstChar = offsetof(StringObject, m_FirstChar);
+    static constexpr size_t m_StringLength = offsetof(StringObject, m_StringLength);
 };
 
 /*================================GetEmptyString================================
diff --git a/src/native/managed/cdacreader/src/Constants.cs b/src/native/managed/cdacreader/src/Constants.cs
@@ -14,9 +14,11 @@ internal static class Globals
         internal const string GCThread = nameof(GCThread);
 
         internal const string FeatureEHFunclets = nameof(FeatureEHFunclets);
+        internal const string ObjectToMethodTableUnmask = nameof(ObjectToMethodTableUnmask);
         internal const string SOSBreakingChangeVersion = nameof(SOSBreakingChangeVersion);
 
         internal const string FreeObjectMethodTable = nameof(FreeObjectMethodTable);
+        internal const string StringMethodTable = nameof(StringMethodTable);
 
         internal const string MiniMetaDataBuffAddress = nameof(MiniMetaDataBuffAddress);
         internal const string MiniMetaDataBuffMaxSize = nameof(MiniMetaDataBuffMaxSize);
diff --git a/src/native/managed/cdacreader/src/Contracts/Object.cs b/src/native/managed/cdacreader/src/Contracts/Object.cs
@@ -0,0 +1,31 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+
+using System;
+
+namespace Microsoft.Diagnostics.DataContractReader.Contracts;
+
+internal interface IObject : IContract
+{
+    static string IContract.Name { get; } = nameof(Object);
+    static IContract IContract.Create(Target target, int version)
+    {
+        ulong methodTableOffset = (ulong)target.GetTypeInfo(DataType.Object).Fields["m_pMethTab"].Offset;
+        byte objectToMethodTableUnmask = target.ReadGlobal<byte>(Constants.Globals.ObjectToMethodTableUnmask);
+        TargetPointer stringMethodTable = target.ReadPointer(
+            target.ReadGlobalPointer(Constants.Globals.StringMethodTable));
+        return version switch
+        {
+            1 => new Object_1(target, methodTableOffset, objectToMethodTableUnmask, stringMethodTable),
+            _ => default(Object),
+        };
+    }
+
+    public virtual TargetPointer GetMethodTableAddress(TargetPointer address) => throw new NotImplementedException();
+    public virtual string GetStringValue(TargetPointer address) => throw new NotImplementedException();
+}
+
+internal readonly struct Object : IObject
+{
+    // Everything throws NotImplementedException
+}
diff --git a/src/native/managed/cdacreader/src/Contracts/Object_1.cs b/src/native/managed/cdacreader/src/Contracts/Object_1.cs
@@ -0,0 +1,41 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+
+using System;
+using System.Runtime.InteropServices;
+
+namespace Microsoft.Diagnostics.DataContractReader.Contracts;
+
+internal readonly struct Object_1 : IObject
+{
+    private readonly Target _target;
+    private readonly ulong _methodTableOffset;
+    private readonly TargetPointer _stringMethodTable;
+    private readonly byte _objectToMethodTableUnmask;
+
+    internal Object_1(Target target, ulong methodTableOffset, byte objectToMethodTableUnmask, TargetPointer stringMethodTable)
+    {
+        _target = target;
+        _methodTableOffset = methodTableOffset;
+        _stringMethodTable = stringMethodTable;
+        _objectToMethodTableUnmask = objectToMethodTableUnmask;
+    }
+
+    public TargetPointer GetMethodTableAddress(TargetPointer address)
+    {
+        TargetPointer mt = _target.ReadPointer(address + _methodTableOffset);
+        return mt.Value & (ulong)~_objectToMethodTableUnmask;
+    }
+
+    string IObject.GetStringValue(TargetPointer address)
+    {
+        TargetPointer mt = GetMethodTableAddress(address);
+        if (mt != _stringMethodTable)
+            throw new ArgumentException("Address does not represent a string object", nameof(address));
+
+        Data.String str = _target.ProcessedData.GetOrAdd<Data.String>(address);
+        Span<byte> span = stackalloc byte[(int)str.StringLength * sizeof(char)];
+        _target.ReadBuffer(str.FirstChar, span);
+        return new string(MemoryMarshal.Cast<byte, char>(span));
+    }
+}
diff --git a/src/native/managed/cdacreader/src/Contracts/Registry.cs b/src/native/managed/cdacreader/src/Contracts/Registry.cs
@@ -20,6 +20,7 @@ public Registry(Target target)
 
     public IException Exception => GetContract<IException>();
     public ILoader Loader => GetContract<ILoader>();
+    public IObject Object => GetContract<IObject>();
     public IThread Thread => GetContract<IThread>();
     public IRuntimeTypeSystem RuntimeTypeSystem => GetContract<IRuntimeTypeSystem>();
     public IDacStreams DacStreams => GetContract<IDacStreams>();
diff --git a/src/native/managed/cdacreader/src/Data/String.cs b/src/native/managed/cdacreader/src/Data/String.cs
@@ -0,0 +1,21 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+
+namespace Microsoft.Diagnostics.DataContractReader.Data;
+
+internal sealed class String : IData<String>
+{
+    static String IData<String>.Create(Target target, TargetPointer address)
+        => new String(target, address);
+
+    public String(Target target, TargetPointer address)
+    {
+        Target.TypeInfo type = target.GetTypeInfo(DataType.String);
+
+        FirstChar = address + (ulong)type.Fields["m_FirstChar"].Offset;
+        StringLength = target.Read<uint>(address + (ulong)type.Fields["m_StringLength"].Offset);
+    }
+
+    public TargetPointer FirstChar { get; init; }
+    public uint StringLength { get; init; }
+}
diff --git a/src/native/managed/cdacreader/src/DataType.cs b/src/native/managed/cdacreader/src/DataType.cs
@@ -37,4 +37,6 @@ public enum DataType
     TypeVarTypeDesc,
     FnPtrTypeDesc,
     DynamicMetadata,
+    Object,
+    String,
 }
diff --git a/src/native/managed/cdacreader/src/Legacy/SOSDacImpl.cs b/src/native/managed/cdacreader/src/Legacy/SOSDacImpl.cs
@@ -308,7 +308,21 @@ public unsafe int GetObjectExceptionData(ulong objectAddress, DacpExceptionObjec
         return HResults.S_OK;
     }
 
-    public unsafe int GetObjectStringData(ulong obj, uint count, char* stringData, uint* pNeeded) => HResults.E_NOTIMPL;
+    public unsafe int GetObjectStringData(ulong obj, uint count, char* stringData, uint* pNeeded)
+    {
+        try
+        {
+            Contracts.IObject contract = _target.Contracts.Object;
+            string str = contract.GetStringValue(obj);
+            CopyStringToTargetBuffer(stringData, count, pNeeded, str);
+        }
+        catch (System.Exception ex)
+        {
+            return ex.HResult;
+        }
+
+        return HResults.S_OK;
+    }
     public unsafe int GetOOMData(ulong oomAddr, void* data) => HResults.E_NOTIMPL;
     public unsafe int GetOOMStaticData(void* data) => HResults.E_NOTIMPL;
     public unsafe int GetPEFileBase(ulong addr, ulong* peBase) => HResults.E_NOTIMPL;
diff --git a/src/native/managed/cdacreader/tests/ObjectTests.cs b/src/native/managed/cdacreader/tests/ObjectTests.cs

Original file line number	Diff line number	Diff line change
`@@ -12,6 +12,7 @@`
`12`	`12`	`"DacStreams": 1,`
`13`	`13`	`"Exception": 1,`
`14`	`14`	`"Loader": 1,`
	`15`	`+ "Object": 1,`
`15`	`16`	`"RuntimeTypeSystem": 1,`
`16`	`17`	`"Thread": 1`
`17`	`18`	`}`
Original file line number	Diff line number	Diff line change
`@@ -37,4 +37,6 @@ public enum DataType`
`37`	`37`	`TypeVarTypeDesc,`
`38`	`38`	`FnPtrTypeDesc,`
`39`	`39`	`DynamicMetadata,`
	`40`	`+ Object,`
	`41`	`+ String,`
`40`	`42`	`}`