llvm-project/lldb/source/Target/ObjCLanguageRuntime.cpp

629 lines
20 KiB
C++
Raw Normal View History

//===-- ObjCLanguageRuntime.cpp ---------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "clang/AST/Type.h"
#include "lldb/Core/Log.h"
#include "lldb/Core/MappedHash.h"
<rdar://problem/11757916> Make breakpoint setting by file and line much more efficient by only looking for inlined breakpoint locations if we are setting a breakpoint in anything but a source implementation file. Implementing this complex for a many reasons. Turns out that parsing compile units lazily had some issues with respect to how we need to do things with DWARF in .o files. So the fixes in the checkin for this makes these changes: - Add a new setting called "target.inline-breakpoint-strategy" which can be set to "never", "always", or "headers". "never" will never try and set any inlined breakpoints (fastest). "always" always looks for inlined breakpoint locations (slowest, but most accurate). "headers", which is the default setting, will only look for inlined breakpoint locations if the breakpoint is set in what are consudered to be header files, which is realy defined as "not in an implementation source file". - modify the breakpoint setting by file and line to check the current "target.inline-breakpoint-strategy" setting and act accordingly - Modify compile units to be able to get their language and other info lazily. This allows us to create compile units from the debug map and not have to fill all of the details in, and then lazily discover this information as we go on debuggging. This is needed to avoid parsing all .o files when setting breakpoints in implementation only files (no inlines). Otherwise we would need to parse the .o file, the object file (mach-o in our case) and the symbol file (DWARF in the object file) just to see what the compile unit was. - modify the "SymbolFileDWARFDebugMap" to subclass lldb_private::Module so that the virtual "GetObjectFile()" and "GetSymbolVendor()" functions can be intercepted when the .o file contenst are later lazilly needed. Prior to this fix, when we first instantiated the "SymbolFileDWARFDebugMap" class, we would also make modules, object files and symbol files for every .o file in the debug map because we needed to fix up the sections in the .o files with information that is in the executable debug map. Now we lazily do this in the DebugMapModule::GetObjectFile() Cleaned up header includes a bit as well. llvm-svn: 162860
2012-08-30 05:13:06 +08:00
#include "lldb/Core/Module.h"
#include "lldb/Core/PluginManager.h"
#include "lldb/Core/Timer.h"
#include "lldb/Core/ValueObject.h"
#include "lldb/Symbol/ClangASTContext.h"
#include "lldb/Symbol/Type.h"
<rdar://problem/11757916> Make breakpoint setting by file and line much more efficient by only looking for inlined breakpoint locations if we are setting a breakpoint in anything but a source implementation file. Implementing this complex for a many reasons. Turns out that parsing compile units lazily had some issues with respect to how we need to do things with DWARF in .o files. So the fixes in the checkin for this makes these changes: - Add a new setting called "target.inline-breakpoint-strategy" which can be set to "never", "always", or "headers". "never" will never try and set any inlined breakpoints (fastest). "always" always looks for inlined breakpoint locations (slowest, but most accurate). "headers", which is the default setting, will only look for inlined breakpoint locations if the breakpoint is set in what are consudered to be header files, which is realy defined as "not in an implementation source file". - modify the breakpoint setting by file and line to check the current "target.inline-breakpoint-strategy" setting and act accordingly - Modify compile units to be able to get their language and other info lazily. This allows us to create compile units from the debug map and not have to fill all of the details in, and then lazily discover this information as we go on debuggging. This is needed to avoid parsing all .o files when setting breakpoints in implementation only files (no inlines). Otherwise we would need to parse the .o file, the object file (mach-o in our case) and the symbol file (DWARF in the object file) just to see what the compile unit was. - modify the "SymbolFileDWARFDebugMap" to subclass lldb_private::Module so that the virtual "GetObjectFile()" and "GetSymbolVendor()" functions can be intercepted when the .o file contenst are later lazilly needed. Prior to this fix, when we first instantiated the "SymbolFileDWARFDebugMap" class, we would also make modules, object files and symbol files for every .o file in the debug map because we needed to fix up the sections in the .o files with information that is in the executable debug map. Now we lazily do this in the DebugMapModule::GetObjectFile() Cleaned up header includes a bit as well. llvm-svn: 162860
2012-08-30 05:13:06 +08:00
#include "lldb/Symbol/TypeList.h"
#include "lldb/Target/ObjCLanguageRuntime.h"
#include "lldb/Target/Target.h"
#include "llvm/ADT/StringRef.h"
using namespace lldb;
using namespace lldb_private;
//----------------------------------------------------------------------
// Destructor
//----------------------------------------------------------------------
ObjCLanguageRuntime::~ObjCLanguageRuntime()
{
}
ObjCLanguageRuntime::ObjCLanguageRuntime (Process *process) :
LanguageRuntime (process),
m_has_new_literals_and_indexing (eLazyBoolCalculate),
m_isa_to_descriptor(),
m_isa_to_descriptor_stop_id (UINT32_MAX)
{
}
bool
ObjCLanguageRuntime::AddClass (ObjCISA isa, const ClassDescriptorSP &descriptor_sp, const char *class_name)
{
if (isa != 0)
{
m_isa_to_descriptor[isa] = descriptor_sp;
// class_name is assumed to be valid
m_hash_to_isa_map.insert(std::make_pair(MappedHash::HashStringUsingDJB(class_name), isa));
return true;
}
return false;
}
void
ObjCLanguageRuntime::AddToMethodCache (lldb::addr_t class_addr, lldb::addr_t selector, lldb::addr_t impl_addr)
{
Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_STEP));
if (log)
{
log->Printf ("Caching: class 0x%" PRIx64 " selector 0x%" PRIx64 " implementation 0x%" PRIx64 ".", class_addr, selector, impl_addr);
}
m_impl_cache.insert (std::pair<ClassAndSel,lldb::addr_t> (ClassAndSel(class_addr, selector), impl_addr));
}
lldb::addr_t
ObjCLanguageRuntime::LookupInMethodCache (lldb::addr_t class_addr, lldb::addr_t selector)
{
MsgImplMap::iterator pos, end = m_impl_cache.end();
pos = m_impl_cache.find (ClassAndSel(class_addr, selector));
if (pos != end)
return (*pos).second;
return LLDB_INVALID_ADDRESS;
}
lldb::TypeSP
ObjCLanguageRuntime::LookupInCompleteClassCache (ConstString &name)
{
CompleteClassMap::iterator complete_class_iter = m_complete_class_cache.find(name);
if (complete_class_iter != m_complete_class_cache.end())
{
// Check the weak pointer to make sure the type hasn't been unloaded
TypeSP complete_type_sp (complete_class_iter->second.lock());
if (complete_type_sp)
return complete_type_sp;
else
m_complete_class_cache.erase(name);
}
if (m_negative_complete_class_cache.count(name) > 0)
return TypeSP();
const ModuleList &modules = m_process->GetTarget().GetImages();
SymbolContextList sc_list;
const size_t matching_symbols = modules.FindSymbolsWithNameAndType (name,
eSymbolTypeObjCClass,
sc_list);
if (matching_symbols)
{
SymbolContext sc;
sc_list.GetContextAtIndex(0, sc);
ModuleSP module_sp(sc.module_sp);
if (!module_sp)
return TypeSP();
const SymbolContext null_sc;
const bool exact_match = true;
const uint32_t max_matches = UINT32_MAX;
TypeList types;
const uint32_t num_types = module_sp->FindTypes (null_sc,
name,
exact_match,
max_matches,
types);
if (num_types)
{
uint32_t i;
for (i = 0; i < num_types; ++i)
{
TypeSP type_sp (types.GetTypeAtIndex(i));
if (type_sp->GetClangForwardType().IsObjCObjectOrInterfaceType())
{
if (type_sp->IsCompleteObjCClass())
{
m_complete_class_cache[name] = type_sp;
return type_sp;
}
}
}
}
}
m_negative_complete_class_cache.insert(name);
return TypeSP();
}
size_t
ObjCLanguageRuntime::GetByteOffsetForIvar (ClangASTType &parent_qual_type, const char *ivar_name)
{
return LLDB_INVALID_IVAR_OFFSET;
}
void
ObjCLanguageRuntime::MethodName::Clear()
{
m_full.Clear();
m_class.Clear();
m_category.Clear();
m_selector.Clear();
m_type = eTypeUnspecified;
m_category_is_valid = false;
}
//bool
//ObjCLanguageRuntime::MethodName::SetName (const char *name, bool strict)
//{
// Clear();
// if (name && name[0])
// {
// // If "strict" is true. then the method must be specified with a
// // '+' or '-' at the beginning. If "strict" is false, then the '+'
// // or '-' can be omitted
// bool valid_prefix = false;
//
// if (name[0] == '+' || name[0] == '-')
// {
// valid_prefix = name[1] == '[';
// }
// else if (!strict)
// {
// // "strict" is false, the name just needs to start with '['
// valid_prefix = name[0] == '[';
// }
//
// if (valid_prefix)
// {
// static RegularExpression g_regex("^([-+]?)\\[([A-Za-z_][A-Za-z_0-9]*)(\\([A-Za-z_][A-Za-z_0-9]*\\))? ([A-Za-z_][A-Za-z_0-9:]*)\\]$");
// llvm::StringRef matches[4];
// // Since we are using a global regular expression, we must use the threadsafe version of execute
// if (g_regex.ExecuteThreadSafe(name, matches, 4))
// {
// m_full.SetCString(name);
// if (matches[0].empty())
// m_type = eTypeUnspecified;
// else if (matches[0][0] == '+')
// m_type = eTypeClassMethod;
// else
// m_type = eTypeInstanceMethod;
// m_class.SetString(matches[1]);
// m_selector.SetString(matches[3]);
// if (!matches[2].empty())
// m_category.SetString(matches[2]);
// }
// }
// }
// return IsValid(strict);
//}
bool
ObjCLanguageRuntime::MethodName::SetName (const char *name, bool strict)
{
Clear();
if (name && name[0])
{
// If "strict" is true. then the method must be specified with a
// '+' or '-' at the beginning. If "strict" is false, then the '+'
// or '-' can be omitted
bool valid_prefix = false;
if (name[0] == '+' || name[0] == '-')
{
valid_prefix = name[1] == '[';
if (name[0] == '+')
m_type = eTypeClassMethod;
else
m_type = eTypeInstanceMethod;
}
else if (!strict)
{
// "strict" is false, the name just needs to start with '['
valid_prefix = name[0] == '[';
}
if (valid_prefix)
{
int name_len = strlen (name);
// Objective C methods must have at least:
// "-[" or "+[" prefix
// One character for a class name
// One character for the space between the class name
// One character for the method name
// "]" suffix
if (name_len >= (5 + (strict ? 1 : 0)) && name[name_len - 1] == ']')
{
m_full.SetCStringWithLength(name, name_len);
}
}
}
return IsValid(strict);
}
const ConstString &
ObjCLanguageRuntime::MethodName::GetClassName ()
{
if (!m_class)
{
if (IsValid(false))
{
const char *full = m_full.GetCString();
const char *class_start = (full[0] == '[' ? full + 1 : full + 2);
const char *paren_pos = strchr (class_start, '(');
if (paren_pos)
{
m_class.SetCStringWithLength (class_start, paren_pos - class_start);
}
else
{
// No '(' was found in the full name, we can definitively say
// that our category was valid (and empty).
m_category_is_valid = true;
const char *space_pos = strchr (full, ' ');
if (space_pos)
{
m_class.SetCStringWithLength (class_start, space_pos - class_start);
if (!m_class_category)
{
// No category in name, so we can also fill in the m_class_category
m_class_category = m_class;
}
}
}
}
}
return m_class;
}
const ConstString &
ObjCLanguageRuntime::MethodName::GetClassNameWithCategory ()
{
if (!m_class_category)
{
if (IsValid(false))
{
const char *full = m_full.GetCString();
const char *class_start = (full[0] == '[' ? full + 1 : full + 2);
const char *space_pos = strchr (full, ' ');
if (space_pos)
{
m_class_category.SetCStringWithLength (class_start, space_pos - class_start);
// If m_class hasn't been filled in and the class with category doesn't
// contain a '(', then we can also fill in the m_class
if (!m_class && strchr (m_class_category.GetCString(), '(') == NULL)
{
m_class = m_class_category;
// No '(' was found in the full name, we can definitively say
// that our category was valid (and empty).
m_category_is_valid = true;
}
}
}
}
return m_class_category;
}
const ConstString &
ObjCLanguageRuntime::MethodName::GetSelector ()
{
if (!m_selector)
{
if (IsValid(false))
{
const char *full = m_full.GetCString();
const char *space_pos = strchr (full, ' ');
if (space_pos)
{
++space_pos; // skip the space
m_selector.SetCStringWithLength (space_pos, m_full.GetLength() - (space_pos - full) - 1);
}
}
}
return m_selector;
}
const ConstString &
ObjCLanguageRuntime::MethodName::GetCategory ()
{
if (!m_category_is_valid && !m_category)
{
if (IsValid(false))
{
m_category_is_valid = true;
const char *full = m_full.GetCString();
const char *class_start = (full[0] == '[' ? full + 1 : full + 2);
const char *open_paren_pos = strchr (class_start, '(');
if (open_paren_pos)
{
++open_paren_pos; // Skip the open paren
const char *close_paren_pos = strchr (open_paren_pos, ')');
if (close_paren_pos)
m_category.SetCStringWithLength (open_paren_pos, close_paren_pos - open_paren_pos);
}
}
}
return m_category;
}
ConstString
ObjCLanguageRuntime::MethodName::GetFullNameWithoutCategory (bool empty_if_no_category)
{
if (IsValid(false))
{
if (HasCategory())
{
StreamString strm;
if (m_type == eTypeClassMethod)
strm.PutChar('+');
else if (m_type == eTypeInstanceMethod)
strm.PutChar('-');
strm.Printf("[%s %s]", GetClassName().GetCString(), GetSelector().GetCString());
return ConstString(strm.GetString().c_str());
}
if (!empty_if_no_category)
{
// Just return the full name since it doesn't have a category
return GetFullName();
}
}
return ConstString();
}
size_t
ObjCLanguageRuntime::MethodName::GetFullNames (std::vector<ConstString> &names, bool append)
{
if (!append)
names.clear();
if (IsValid(false))
{
StreamString strm;
const bool is_class_method = m_type == eTypeClassMethod;
const bool is_instance_method = m_type == eTypeInstanceMethod;
const ConstString &category = GetCategory();
if (is_class_method || is_instance_method)
{
names.push_back (m_full);
if (category)
{
strm.Printf("%c[%s %s]",
is_class_method ? '+' : '-',
GetClassName().GetCString(),
GetSelector().GetCString());
names.push_back(ConstString(strm.GetString().c_str()));
}
}
else
{
const ConstString &class_name = GetClassName();
const ConstString &selector = GetSelector();
strm.Printf("+[%s %s]", class_name.GetCString(), selector.GetCString());
names.push_back(ConstString(strm.GetString().c_str()));
strm.Clear();
strm.Printf("-[%s %s]", class_name.GetCString(), selector.GetCString());
names.push_back(ConstString(strm.GetString().c_str()));
strm.Clear();
if (category)
{
strm.Printf("+[%s(%s) %s]", class_name.GetCString(), category.GetCString(), selector.GetCString());
names.push_back(ConstString(strm.GetString().c_str()));
strm.Clear();
strm.Printf("-[%s(%s) %s]", class_name.GetCString(), category.GetCString(), selector.GetCString());
names.push_back(ConstString(strm.GetString().c_str()));
}
}
}
return names.size();
}
bool
ObjCLanguageRuntime::ClassDescriptor::IsPointerValid (lldb::addr_t value,
uint32_t ptr_size,
bool allow_NULLs,
bool allow_tagged,
bool check_version_specific) const
{
if (!value)
return allow_NULLs;
if ( (value % 2) == 1 && allow_tagged)
return true;
if ((value % ptr_size) == 0)
return (check_version_specific ? CheckPointer(value,ptr_size) : true);
else
return false;
}
ObjCLanguageRuntime::ObjCISA
ObjCLanguageRuntime::GetISA(const ConstString &name)
{
ISAToDescriptorIterator pos = GetDescriptorIterator (name);
if (pos != m_isa_to_descriptor.end())
return pos->first;
return 0;
}
ObjCLanguageRuntime::ISAToDescriptorIterator
ObjCLanguageRuntime::GetDescriptorIterator (const ConstString &name)
{
ISAToDescriptorIterator end = m_isa_to_descriptor.end();
if (name)
{
UpdateISAToDescriptorMap();
if (m_hash_to_isa_map.empty())
{
// No name hashes were provided, we need to just linearly power through the
// names and find a match
for (ISAToDescriptorIterator pos = m_isa_to_descriptor.begin(); pos != end; ++pos)
{
if (pos->second->GetClassName() == name)
return pos;
}
}
else
{
// Name hashes were provided, so use them to efficiently lookup name to isa/descriptor
const uint32_t name_hash = MappedHash::HashStringUsingDJB (name.GetCString());
std::pair <HashToISAIterator, HashToISAIterator> range = m_hash_to_isa_map.equal_range(name_hash);
for (HashToISAIterator range_pos = range.first; range_pos != range.second; ++range_pos)
{
ISAToDescriptorIterator pos = m_isa_to_descriptor.find (range_pos->second);
if (pos != m_isa_to_descriptor.end())
{
if (pos->second->GetClassName() == name)
return pos;
}
}
}
}
return end;
}
ObjCLanguageRuntime::ObjCISA
ObjCLanguageRuntime::GetParentClass(ObjCLanguageRuntime::ObjCISA isa)
{
ClassDescriptorSP objc_class_sp (GetClassDescriptorFromISA(isa));
if (objc_class_sp)
{
ClassDescriptorSP objc_super_class_sp (objc_class_sp->GetSuperclass());
if (objc_super_class_sp)
return objc_super_class_sp->GetISA();
}
return 0;
}
ConstString
ObjCLanguageRuntime::GetActualTypeName(ObjCLanguageRuntime::ObjCISA isa)
{
ClassDescriptorSP objc_class_sp (GetNonKVOClassDescriptor(isa));
if (objc_class_sp)
return objc_class_sp->GetClassName();
return ConstString();
}
ObjCLanguageRuntime::ClassDescriptorSP
ObjCLanguageRuntime::GetClassDescriptorFromClassName (const ConstString &class_name)
{
ISAToDescriptorIterator pos = GetDescriptorIterator (class_name);
if (pos != m_isa_to_descriptor.end())
return pos->second;
return ClassDescriptorSP();
}
ObjCLanguageRuntime::ClassDescriptorSP
ObjCLanguageRuntime::GetClassDescriptor (ValueObject& valobj)
{
ClassDescriptorSP objc_class_sp;
// if we get an invalid VO (which might still happen when playing around
// with pointers returned by the expression parser, don't consider this
// a valid ObjC object)
if (valobj.GetClangType().IsValid())
{
addr_t isa_pointer = valobj.GetPointerValue();
if (isa_pointer != LLDB_INVALID_ADDRESS)
{
ExecutionContext exe_ctx (valobj.GetExecutionContextRef());
Process *process = exe_ctx.GetProcessPtr();
if (process)
{
Error error;
ObjCISA isa = process->ReadPointerFromMemory(isa_pointer, error);
if (isa != LLDB_INVALID_ADDRESS)
objc_class_sp = GetClassDescriptorFromISA (isa);
}
}
}
return objc_class_sp;
}
ObjCLanguageRuntime::ClassDescriptorSP
ObjCLanguageRuntime::GetNonKVOClassDescriptor (ValueObject& valobj)
{
ObjCLanguageRuntime::ClassDescriptorSP objc_class_sp (GetClassDescriptor (valobj));
if (objc_class_sp)
{
if (!objc_class_sp->IsKVO())
return objc_class_sp;
ClassDescriptorSP non_kvo_objc_class_sp(objc_class_sp->GetSuperclass());
if (non_kvo_objc_class_sp && non_kvo_objc_class_sp->IsValid())
return non_kvo_objc_class_sp;
}
return ClassDescriptorSP();
}
ObjCLanguageRuntime::ClassDescriptorSP
ObjCLanguageRuntime::GetClassDescriptorFromISA (ObjCISA isa)
{
if (isa)
{
UpdateISAToDescriptorMap();
ObjCLanguageRuntime::ISAToDescriptorIterator pos = m_isa_to_descriptor.find(isa);
if (pos != m_isa_to_descriptor.end())
return pos->second;
}
return ClassDescriptorSP();
}
ObjCLanguageRuntime::ClassDescriptorSP
ObjCLanguageRuntime::GetNonKVOClassDescriptor (ObjCISA isa)
{
if (isa)
{
ClassDescriptorSP objc_class_sp = GetClassDescriptorFromISA (isa);
if (objc_class_sp && objc_class_sp->IsValid())
{
if (!objc_class_sp->IsKVO())
return objc_class_sp;
ClassDescriptorSP non_kvo_objc_class_sp(objc_class_sp->GetSuperclass());
if (non_kvo_objc_class_sp && non_kvo_objc_class_sp->IsValid())
return non_kvo_objc_class_sp;
}
}
return ClassDescriptorSP();
}
This is a fairly bulky patch, but a lot of it involves rearranging existing code What it does: - it introduces a concept of EncodingToType to the ObjCLanguageRuntime The ObjC runtime has a "type encoding" feature that describes types as strings The EncodingToType is a decoder for that format, making types out of type encoding strings This feature already existed in some shape as we were using it to create method signatures out of the runtime, but this checkin extends the parser to support the full syntax, and moves things so that more parts of LLDB have access to this decoder - it splits the ClassDescriptorV2 object to its own file, it was starting to grow too large - it adds to the ClassDescriptor mechanism a notion of ivar storage; the ObjC runtime vends ivar information as well as method information While ivar information is not ready for prime type (i.e. we don't want to add it to the runtime generated types for expression evaluator usage), there are potentially useful scenarios in which realizing ivar types could be useful. For now, the ClassDescriptor is going to hold ivar information directly. Existing code already allows describing ivars, this patch hooks those moving parts up so that one can actually ask a ClassDescriptor about ivars for the class it represents and as a couple minor niceties: - it makes it possible to retrieve the LLDB ClangASTContext that is associated to a clang::ASTContext - it extends the ValueObject-to-ClassDescriptor API in the language runtime to deal correctly with base-class hierarchies llvm-svn: 216026
2014-08-20 05:46:37 +08:00
ClangASTType
ObjCLanguageRuntime::EncodingToType::RealizeType (const char* name, bool allow_unknownanytype)
{
if (m_scratch_ast_ctx_ap)
return RealizeType(*m_scratch_ast_ctx_ap, name, allow_unknownanytype);
return ClangASTType();
}
ClangASTType
ObjCLanguageRuntime::EncodingToType::RealizeType (ClangASTContext& ast_ctx, const char* name, bool allow_unknownanytype)
{
clang::ASTContext *clang_ast = ast_ctx.getASTContext();
if (!clang_ast)
return ClangASTType();
return RealizeType(*clang_ast, name, allow_unknownanytype);
}
ObjCLanguageRuntime::EncodingToType::~EncodingToType() {}
This is a fairly bulky patch, but a lot of it involves rearranging existing code What it does: - it introduces a concept of EncodingToType to the ObjCLanguageRuntime The ObjC runtime has a "type encoding" feature that describes types as strings The EncodingToType is a decoder for that format, making types out of type encoding strings This feature already existed in some shape as we were using it to create method signatures out of the runtime, but this checkin extends the parser to support the full syntax, and moves things so that more parts of LLDB have access to this decoder - it splits the ClassDescriptorV2 object to its own file, it was starting to grow too large - it adds to the ClassDescriptor mechanism a notion of ivar storage; the ObjC runtime vends ivar information as well as method information While ivar information is not ready for prime type (i.e. we don't want to add it to the runtime generated types for expression evaluator usage), there are potentially useful scenarios in which realizing ivar types could be useful. For now, the ClassDescriptor is going to hold ivar information directly. Existing code already allows describing ivars, this patch hooks those moving parts up so that one can actually ask a ClassDescriptor about ivars for the class it represents and as a couple minor niceties: - it makes it possible to retrieve the LLDB ClangASTContext that is associated to a clang::ASTContext - it extends the ValueObject-to-ClassDescriptor API in the language runtime to deal correctly with base-class hierarchies llvm-svn: 216026
2014-08-20 05:46:37 +08:00
ObjCLanguageRuntime::EncodingToTypeSP
ObjCLanguageRuntime::GetEncodingToType ()
{
return nullptr;
}