llvm-project/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugLine.cpp

1427 lines
56 KiB
C++
Raw Normal View History

//===-- DWARFDebugLine.cpp --------------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "DWARFDebugLine.h"
//#define ENABLE_DEBUG_PRINTF // DO NOT LEAVE THIS DEFINED: DEBUG ONLY!!!
#include <assert.h>
#include "lldb/Core/FileSpecList.h"
#include "lldb/Core/Log.h"
#include "lldb/Core/Timer.h"
#include "lldb/Host/Host.h"
#include "SymbolFileDWARF.h"
#include "LogChannelDWARF.h"
using namespace lldb;
using namespace lldb_private;
using namespace std;
//----------------------------------------------------------------------
// Parse
//
// Parse all information in the debug_line_data into an internal
// representation.
//----------------------------------------------------------------------
void
DWARFDebugLine::Parse(const DataExtractor& debug_line_data)
{
m_lineTableMap.clear();
dw_offset_t offset = 0;
LineTable::shared_ptr line_table_sp(new LineTable);
while (debug_line_data.ValidOffset(offset))
{
const uint32_t debug_line_offset = offset;
if (line_table_sp.get() == NULL)
break;
if (ParseStatementTable(debug_line_data, &offset, line_table_sp.get()))
{
// Make sure we don't don't loop infinitely
if (offset <= debug_line_offset)
break;
//DEBUG_PRINTF("m_lineTableMap[0x%8.8x] = line_table_sp\n", debug_line_offset);
m_lineTableMap[debug_line_offset] = line_table_sp;
line_table_sp.reset(new LineTable);
}
else
++offset; // Try next byte in line table
}
}
void
DWARFDebugLine::ParseIfNeeded(const DataExtractor& debug_line_data)
{
if (m_lineTableMap.empty())
Parse(debug_line_data);
}
//----------------------------------------------------------------------
// DWARFDebugLine::GetLineTable
//----------------------------------------------------------------------
DWARFDebugLine::LineTable::shared_ptr
DWARFDebugLine::GetLineTable(const dw_offset_t offset) const
{
DWARFDebugLine::LineTable::shared_ptr line_table_shared_ptr;
LineTableConstIter pos = m_lineTableMap.find(offset);
if (pos != m_lineTableMap.end())
line_table_shared_ptr = pos->second;
return line_table_shared_ptr;
}
//----------------------------------------------------------------------
// DumpStateToFile
//----------------------------------------------------------------------
static void
DumpStateToFile (dw_offset_t offset, const DWARFDebugLine::State& state, void* userData)
{
Log *log = (Log *)userData;
if (state.row == DWARFDebugLine::State::StartParsingLineTable)
{
// If the row is zero we are being called with the prologue only
state.prologue->Dump (log);
log->PutCString ("Address Line Column File");
log->PutCString ("------------------ ------ ------ ------");
}
else if (state.row == DWARFDebugLine::State::DoneParsingLineTable)
{
// Done parsing line table
}
else
{
log->Printf( "0x%16.16llx %6u %6u %6u%s\n", state.address, state.line, state.column, state.file, state.end_sequence ? " END" : "");
}
}
//----------------------------------------------------------------------
// DWARFDebugLine::DumpLineTableRows
//----------------------------------------------------------------------
bool
DWARFDebugLine::DumpLineTableRows(Log *log, SymbolFileDWARF* dwarf2Data, dw_offset_t debug_line_offset)
{
const DataExtractor& debug_line_data = dwarf2Data->get_debug_line_data();
if (debug_line_offset == DW_INVALID_OFFSET)
{
// Dump line table to a single file only
debug_line_offset = 0;
while (debug_line_data.ValidOffset(debug_line_offset))
debug_line_offset = DumpStatementTable (log, debug_line_data, debug_line_offset);
}
else
{
// Dump line table to a single file only
DumpStatementTable (log, debug_line_data, debug_line_offset);
}
return false;
}
//----------------------------------------------------------------------
// DWARFDebugLine::DumpStatementTable
//----------------------------------------------------------------------
dw_offset_t
DWARFDebugLine::DumpStatementTable(Log *log, const DataExtractor& debug_line_data, const dw_offset_t debug_line_offset)
{
if (debug_line_data.ValidOffset(debug_line_offset))
{
uint32_t offset = debug_line_offset;
log->Printf( "----------------------------------------------------------------------\n"
"debug_line[0x%8.8x]\n"
"----------------------------------------------------------------------\n", debug_line_offset);
if (ParseStatementTable(debug_line_data, &offset, DumpStateToFile, log))
return offset;
else
return debug_line_offset + 1; // Skip to next byte in .debug_line section
}
return DW_INVALID_OFFSET;
}
//----------------------------------------------------------------------
// DumpOpcodes
//----------------------------------------------------------------------
bool
DWARFDebugLine::DumpOpcodes(Log *log, SymbolFileDWARF* dwarf2Data, dw_offset_t debug_line_offset, uint32_t dump_flags)
{
const DataExtractor& debug_line_data = dwarf2Data->get_debug_line_data();
if (debug_line_data.GetByteSize() == 0)
{
log->Printf( "< EMPTY >\n");
return false;
}
if (debug_line_offset == DW_INVALID_OFFSET)
{
// Dump line table to a single file only
debug_line_offset = 0;
while (debug_line_data.ValidOffset(debug_line_offset))
debug_line_offset = DumpStatementOpcodes (log, debug_line_data, debug_line_offset, dump_flags);
}
else
{
// Dump line table to a single file only
DumpStatementOpcodes (log, debug_line_data, debug_line_offset, dump_flags);
}
return false;
}
//----------------------------------------------------------------------
// DumpStatementOpcodes
//----------------------------------------------------------------------
dw_offset_t
DWARFDebugLine::DumpStatementOpcodes(Log *log, const DataExtractor& debug_line_data, const dw_offset_t debug_line_offset, uint32_t flags)
{
uint32_t offset = debug_line_offset;
if (debug_line_data.ValidOffset(offset))
{
Prologue prologue;
if (ParsePrologue(debug_line_data, &offset, &prologue))
{
log->PutCString ("----------------------------------------------------------------------");
log->Printf ("debug_line[0x%8.8x]", debug_line_offset);
log->PutCString ("----------------------------------------------------------------------\n");
prologue.Dump (log);
}
else
{
offset = debug_line_offset;
log->Printf( "0x%8.8x: skipping pad byte %2.2x", offset, debug_line_data.GetU8(&offset));
return offset;
}
Row row(prologue.default_is_stmt);
const dw_offset_t end_offset = debug_line_offset + prologue.total_length + sizeof(prologue.total_length);
assert(debug_line_data.ValidOffset(end_offset-1));
while (offset < end_offset)
{
const uint32_t op_offset = offset;
uint8_t opcode = debug_line_data.GetU8(&offset);
switch (opcode)
{
case 0: // Extended Opcodes always start with a zero opcode followed by
{ // a uleb128 length so you can skip ones you don't know about
dw_offset_t ext_offset = offset;
dw_uleb128_t len = debug_line_data.GetULEB128(&offset);
dw_offset_t arg_size = len - (offset - ext_offset);
uint8_t sub_opcode = debug_line_data.GetU8(&offset);
// if (verbose)
// log->Printf( "Extended: <%u> %2.2x ", len, sub_opcode);
switch (sub_opcode)
{
case DW_LNE_end_sequence :
log->Printf( "0x%8.8x: DW_LNE_end_sequence", op_offset);
row.Dump(log);
row.Reset(prologue.default_is_stmt);
break;
case DW_LNE_set_address :
{
row.address = debug_line_data.GetMaxU64(&offset, arg_size);
log->Printf( "0x%8.8x: DW_LNE_set_address (0x%llx)", op_offset, row.address);
}
break;
case DW_LNE_define_file:
{
FileNameEntry fileEntry;
fileEntry.name = debug_line_data.GetCStr(&offset);
fileEntry.dir_idx = debug_line_data.GetULEB128(&offset);
fileEntry.mod_time = debug_line_data.GetULEB128(&offset);
fileEntry.length = debug_line_data.GetULEB128(&offset);
log->Printf( "0x%8.8x: DW_LNE_define_file('%s', dir=%i, mod_time=0x%8.8x, length=%i )",
op_offset,
fileEntry.name.c_str(),
fileEntry.dir_idx,
fileEntry.mod_time,
fileEntry.length);
prologue.file_names.push_back(fileEntry);
}
break;
default:
log->Printf( "0x%8.8x: DW_LNE_??? (%2.2x) - Skipping unknown upcode", op_offset, opcode);
// Length doesn't include the zero opcode byte or the length itself, but
// it does include the sub_opcode, so we have to adjust for that below
offset += arg_size;
break;
}
}
break;
// Standard Opcodes
case DW_LNS_copy:
log->Printf( "0x%8.8x: DW_LNS_copy", op_offset);
row.Dump (log);
break;
case DW_LNS_advance_pc:
{
dw_uleb128_t addr_offset_n = debug_line_data.GetULEB128(&offset);
dw_uleb128_t addr_offset = addr_offset_n * prologue.min_inst_length;
log->Printf( "0x%8.8x: DW_LNS_advance_pc (0x%x)", op_offset, addr_offset);
row.address += addr_offset;
}
break;
case DW_LNS_advance_line:
{
dw_sleb128_t line_offset = debug_line_data.GetSLEB128(&offset);
log->Printf( "0x%8.8x: DW_LNS_advance_line (%i)", op_offset, line_offset);
row.line += line_offset;
}
break;
case DW_LNS_set_file:
row.file = debug_line_data.GetULEB128(&offset);
log->Printf( "0x%8.8x: DW_LNS_set_file (%u)", op_offset, row.file);
break;
case DW_LNS_set_column:
row.column = debug_line_data.GetULEB128(&offset);
log->Printf( "0x%8.8x: DW_LNS_set_column (%u)", op_offset, row.column);
break;
case DW_LNS_negate_stmt:
row.is_stmt = !row.is_stmt;
log->Printf( "0x%8.8x: DW_LNS_negate_stmt", op_offset);
break;
case DW_LNS_set_basic_block:
row.basic_block = true;
log->Printf( "0x%8.8x: DW_LNS_set_basic_block", op_offset);
break;
case DW_LNS_const_add_pc:
{
uint8_t adjust_opcode = 255 - prologue.opcode_base;
dw_addr_t addr_offset = (adjust_opcode / prologue.line_range) * prologue.min_inst_length;
log->Printf( "0x%8.8x: DW_LNS_const_add_pc (0x%8.8llx)", op_offset, addr_offset);
row.address += addr_offset;
}
break;
case DW_LNS_fixed_advance_pc:
{
uint16_t pc_offset = debug_line_data.GetU16(&offset);
log->Printf( "0x%8.8x: DW_LNS_fixed_advance_pc (0x%4.4x)", op_offset, pc_offset);
row.address += pc_offset;
}
break;
case DW_LNS_set_prologue_end:
row.prologue_end = true;
log->Printf( "0x%8.8x: DW_LNS_set_prologue_end", op_offset);
break;
case DW_LNS_set_epilogue_begin:
row.epilogue_begin = true;
log->Printf( "0x%8.8x: DW_LNS_set_epilogue_begin", op_offset);
break;
case DW_LNS_set_isa:
row.isa = debug_line_data.GetULEB128(&offset);
log->Printf( "0x%8.8x: DW_LNS_set_isa (%u)", op_offset, row.isa);
break;
// Special Opcodes
default:
if (opcode < prologue.opcode_base)
{
// We have an opcode that this parser doesn't know about, skip
// the number of ULEB128 numbers that is says to skip in the
// prologue's standard_opcode_lengths array
uint8_t n = prologue.standard_opcode_lengths[opcode-1];
log->Printf( "0x%8.8x: Special : Unknown skipping %u ULEB128 values.", op_offset, n);
while (n > 0)
{
debug_line_data.GetULEB128(&offset);
--n;
}
}
else
{
uint8_t adjust_opcode = opcode - prologue.opcode_base;
dw_addr_t addr_offset = (adjust_opcode / prologue.line_range) * prologue.min_inst_length;
int32_t line_offset = prologue.line_base + (adjust_opcode % prologue.line_range);
log->Printf("0x%8.8x: address += 0x%llx, line += %i\n", op_offset, (uint64_t)addr_offset, line_offset);
row.address += addr_offset;
row.line += line_offset;
row.Dump (log);
}
break;
}
}
return end_offset;
}
return DW_INVALID_OFFSET;
}
//----------------------------------------------------------------------
// Parse
//
// Parse the entire line table contents calling callback each time a
// new prologue is parsed and every time a new row is to be added to
// the line table.
//----------------------------------------------------------------------
void
DWARFDebugLine::Parse(const DataExtractor& debug_line_data, DWARFDebugLine::State::Callback callback, void* userData)
{
uint32_t offset = 0;
if (debug_line_data.ValidOffset(offset))
{
if (!ParseStatementTable(debug_line_data, &offset, callback, userData))
++offset; // Skip to next byte in .debug_line section
}
}
//----------------------------------------------------------------------
// DWARFDebugLine::ParsePrologue
//----------------------------------------------------------------------
bool
DWARFDebugLine::ParsePrologue(const DataExtractor& debug_line_data, dw_offset_t* offset_ptr, Prologue* prologue)
{
const uint32_t prologue_offset = *offset_ptr;
//DEBUG_PRINTF("0x%8.8x: ParsePrologue()\n", *offset_ptr);
prologue->Clear();
uint32_t i;
const char * s;
prologue->total_length = debug_line_data.GetU32(offset_ptr);
prologue->version = debug_line_data.GetU16(offset_ptr);
if (prologue->version != 2)
return false;
prologue->prologue_length = debug_line_data.GetU32(offset_ptr);
const dw_offset_t end_prologue_offset = prologue->prologue_length + *offset_ptr;
prologue->min_inst_length = debug_line_data.GetU8(offset_ptr);
prologue->default_is_stmt = debug_line_data.GetU8(offset_ptr);
prologue->line_base = debug_line_data.GetU8(offset_ptr);
prologue->line_range = debug_line_data.GetU8(offset_ptr);
prologue->opcode_base = debug_line_data.GetU8(offset_ptr);
prologue->standard_opcode_lengths.reserve(prologue->opcode_base-1);
for (i=1; i<prologue->opcode_base; ++i)
{
uint8_t op_len = debug_line_data.GetU8(offset_ptr);
prologue->standard_opcode_lengths.push_back(op_len);
}
while (*offset_ptr < end_prologue_offset)
{
s = debug_line_data.GetCStr(offset_ptr);
if (s && s[0])
prologue->include_directories.push_back(s);
else
break;
}
while (*offset_ptr < end_prologue_offset)
{
const char* name = debug_line_data.GetCStr( offset_ptr );
if (name && name[0])
{
FileNameEntry fileEntry;
fileEntry.name = name;
fileEntry.dir_idx = debug_line_data.GetULEB128( offset_ptr );
fileEntry.mod_time = debug_line_data.GetULEB128( offset_ptr );
fileEntry.length = debug_line_data.GetULEB128( offset_ptr );
prologue->file_names.push_back(fileEntry);
}
else
break;
}
if (*offset_ptr != end_prologue_offset)
{
Host::SystemLog (Host::eSystemLogWarning,
"warning: parsing line table prologue at 0x%8.8x should have ended at 0x%8.8x but it ended ad 0x%8.8x\n",
prologue_offset,
end_prologue_offset,
*offset_ptr);
}
return end_prologue_offset;
}
bool
DWARFDebugLine::ParseSupportFiles(const DataExtractor& debug_line_data, const char *cu_comp_dir, dw_offset_t stmt_list, FileSpecList &support_files)
{
uint32_t offset = stmt_list + 4; // Skip the total length
const char * s;
uint32_t version = debug_line_data.GetU16(&offset);
if (version != 2)
return false;
const dw_offset_t end_prologue_offset = debug_line_data.GetU32(&offset) + offset;
// Skip instruction length, default is stmt, line base, line range and
// opcode base, and all opcode lengths
offset += 4;
const uint8_t opcode_base = debug_line_data.GetU8(&offset);
offset += opcode_base - 1;
std::vector<std::string> include_directories;
include_directories.push_back(""); // Directory at index zero doesn't exist
while (offset < end_prologue_offset)
{
s = debug_line_data.GetCStr(&offset);
if (s && s[0])
include_directories.push_back(s);
else
break;
}
std::string fullpath;
while (offset < end_prologue_offset)
{
const char* path = debug_line_data.GetCStr( &offset );
if (path && path[0])
{
uint32_t dir_idx = debug_line_data.GetULEB128( &offset );
debug_line_data.Skip_LEB128(&offset); // Skip mod_time
debug_line_data.Skip_LEB128(&offset); // Skip length
if (path[0] == '/')
{
// The path starts with a directory delimiter, so we are done.
fullpath = path;
}
else
{
if (dir_idx > 0 && dir_idx < include_directories.size())
{
if (cu_comp_dir && include_directories[dir_idx][0] != '/')
{
fullpath = cu_comp_dir;
if (*fullpath.rbegin() != '/')
fullpath += '/';
fullpath += include_directories[dir_idx];
}
else
fullpath = include_directories[dir_idx];
}
else if (cu_comp_dir && cu_comp_dir[0])
{
fullpath = cu_comp_dir;
}
if (!fullpath.empty())
{
if (*fullpath.rbegin() != '/')
fullpath += '/';
}
fullpath += path;
}
// We don't need to realpath files in the debug_line tables.
FileSpec file_spec(fullpath.c_str(), false);
support_files.Append(file_spec);
}
}
if (offset != end_prologue_offset)
{
Host::SystemLog (Host::eSystemLogError,
"warning: parsing line table prologue at 0x%8.8x should have ended at 0x%8.8x but it ended ad 0x%8.8x\n",
stmt_list,
end_prologue_offset,
offset);
}
return end_prologue_offset;
}
//----------------------------------------------------------------------
// ParseStatementTable
//
// Parse a single line table (prologue and all rows) and call the
// callback function once for the prologue (row in state will be zero)
// and each time a row is to be added to the line table.
//----------------------------------------------------------------------
bool
DWARFDebugLine::ParseStatementTable
(
const DataExtractor& debug_line_data,
dw_offset_t* offset_ptr,
DWARFDebugLine::State::Callback callback,
void* userData
)
{
LogSP log (LogChannelDWARF::GetLogIfAll(DWARF_LOG_DEBUG_LINE));
Prologue::shared_ptr prologue(new Prologue());
const dw_offset_t debug_line_offset = *offset_ptr;
Timer scoped_timer (__PRETTY_FUNCTION__,
"DWARFDebugLine::ParseStatementTable (.debug_line[0x%8.8x])",
debug_line_offset);
if (!ParsePrologue(debug_line_data, offset_ptr, prologue.get()))
{
if (log)
log->Error ("failed to parse DWARF line table prologue");
// Restore our offset and return false to indicate failure!
*offset_ptr = debug_line_offset;
return false;
}
if (log)
prologue->Dump (log.get());
const dw_offset_t end_offset = debug_line_offset + prologue->total_length + sizeof(prologue->total_length);
State state(prologue, log.get(), callback, userData);
while (*offset_ptr < end_offset)
{
//DEBUG_PRINTF("0x%8.8x: ", *offset_ptr);
uint8_t opcode = debug_line_data.GetU8(offset_ptr);
if (opcode == 0)
{
// Extended Opcodes always start with a zero opcode followed by
// a uleb128 length so you can skip ones you don't know about
dw_offset_t ext_offset = *offset_ptr;
dw_uleb128_t len = debug_line_data.GetULEB128(offset_ptr);
dw_offset_t arg_size = len - (*offset_ptr - ext_offset);
//DEBUG_PRINTF("Extended: <%2u> ", len);
uint8_t sub_opcode = debug_line_data.GetU8(offset_ptr);
switch (sub_opcode)
{
case DW_LNE_end_sequence:
// Set the end_sequence register of the state machine to true and
// append a row to the matrix using the current values of the
// state-machine registers. Then reset the registers to the initial
// values specified above. Every statement program sequence must end
// with a DW_LNE_end_sequence instruction which creates a row whose
// address is that of the byte after the last target machine instruction
// of the sequence.
state.end_sequence = true;
state.AppendRowToMatrix(*offset_ptr);
state.Reset();
break;
case DW_LNE_set_address:
// Takes a single relocatable address as an operand. The size of the
// operand is the size appropriate to hold an address on the target
// machine. Set the address register to the value given by the
// relocatable address. All of the other statement program opcodes
// that affect the address register add a delta to it. This instruction
// stores a relocatable value into it instead.
state.address = debug_line_data.GetAddress(offset_ptr);
break;
case DW_LNE_define_file:
// Takes 4 arguments. The first is a null terminated string containing
// a source file name. The second is an unsigned LEB128 number representing
// the directory index of the directory in which the file was found. The
// third is an unsigned LEB128 number representing the time of last
// modification of the file. The fourth is an unsigned LEB128 number
// representing the length in bytes of the file. The time and length
// fields may contain LEB128(0) if the information is not available.
//
// The directory index represents an entry in the include_directories
// section of the statement program prologue. The index is LEB128(0)
// if the file was found in the current directory of the compilation,
// LEB128(1) if it was found in the first directory in the
// include_directories section, and so on. The directory index is
// ignored for file names that represent full path names.
//
// The files are numbered, starting at 1, in the order in which they
// appear; the names in the prologue come before names defined by
// the DW_LNE_define_file instruction. These numbers are used in the
// the file register of the state machine.
{
FileNameEntry fileEntry;
fileEntry.name = debug_line_data.GetCStr(offset_ptr);
fileEntry.dir_idx = debug_line_data.GetULEB128(offset_ptr);
fileEntry.mod_time = debug_line_data.GetULEB128(offset_ptr);
fileEntry.length = debug_line_data.GetULEB128(offset_ptr);
state.prologue->file_names.push_back(fileEntry);
}
break;
default:
// Length doesn't include the zero opcode byte or the length itself, but
// it does include the sub_opcode, so we have to adjust for that below
(*offset_ptr) += arg_size;
break;
}
}
else if (opcode < prologue->opcode_base)
{
switch (opcode)
{
// Standard Opcodes
case DW_LNS_copy:
// Takes no arguments. Append a row to the matrix using the
// current values of the state-machine registers. Then set
// the basic_block register to false.
state.AppendRowToMatrix(*offset_ptr);
break;
case DW_LNS_advance_pc:
// Takes a single unsigned LEB128 operand, multiplies it by the
// min_inst_length field of the prologue, and adds the
// result to the address register of the state machine.
state.address += debug_line_data.GetULEB128(offset_ptr) * prologue->min_inst_length;
break;
case DW_LNS_advance_line:
// Takes a single signed LEB128 operand and adds that value to
// the line register of the state machine.
state.line += debug_line_data.GetSLEB128(offset_ptr);
break;
case DW_LNS_set_file:
// Takes a single unsigned LEB128 operand and stores it in the file
// register of the state machine.
state.file = debug_line_data.GetULEB128(offset_ptr);
break;
case DW_LNS_set_column:
// Takes a single unsigned LEB128 operand and stores it in the
// column register of the state machine.
state.column = debug_line_data.GetULEB128(offset_ptr);
break;
case DW_LNS_negate_stmt:
// Takes no arguments. Set the is_stmt register of the state
// machine to the logical negation of its current value.
state.is_stmt = !state.is_stmt;
break;
case DW_LNS_set_basic_block:
// Takes no arguments. Set the basic_block register of the
// state machine to true
state.basic_block = true;
break;
case DW_LNS_const_add_pc:
// Takes no arguments. Add to the address register of the state
// machine the address increment value corresponding to special
// opcode 255. The motivation for DW_LNS_const_add_pc is this:
// when the statement program needs to advance the address by a
// small amount, it can use a single special opcode, which occupies
// a single byte. When it needs to advance the address by up to
// twice the range of the last special opcode, it can use
// DW_LNS_const_add_pc followed by a special opcode, for a total
// of two bytes. Only if it needs to advance the address by more
// than twice that range will it need to use both DW_LNS_advance_pc
// and a special opcode, requiring three or more bytes.
{
uint8_t adjust_opcode = 255 - prologue->opcode_base;
dw_addr_t addr_offset = (adjust_opcode / prologue->line_range) * prologue->min_inst_length;
state.address += addr_offset;
}
break;
case DW_LNS_fixed_advance_pc:
// Takes a single uhalf operand. Add to the address register of
// the state machine the value of the (unencoded) operand. This
// is the only extended opcode that takes an argument that is not
// a variable length number. The motivation for DW_LNS_fixed_advance_pc
// is this: existing assemblers cannot emit DW_LNS_advance_pc or
// special opcodes because they cannot encode LEB128 numbers or
// judge when the computation of a special opcode overflows and
// requires the use of DW_LNS_advance_pc. Such assemblers, however,
// can use DW_LNS_fixed_advance_pc instead, sacrificing compression.
state.address += debug_line_data.GetU16(offset_ptr);
break;
case DW_LNS_set_prologue_end:
// Takes no arguments. Set the prologue_end register of the
// state machine to true
state.prologue_end = true;
break;
case DW_LNS_set_epilogue_begin:
// Takes no arguments. Set the basic_block register of the
// state machine to true
state.epilogue_begin = true;
break;
case DW_LNS_set_isa:
// Takes a single unsigned LEB128 operand and stores it in the
// column register of the state machine.
state.isa = debug_line_data.GetULEB128(offset_ptr);
break;
default:
// Handle any unknown standard opcodes here. We know the lengths
// of such opcodes because they are specified in the prologue
// as a multiple of LEB128 operands for each opcode.
{
uint8_t i;
assert (opcode - 1 < prologue->standard_opcode_lengths.size());
const uint8_t opcode_length = prologue->standard_opcode_lengths[opcode - 1];
for (i=0; i<opcode_length; ++i)
debug_line_data.Skip_LEB128(offset_ptr);
}
break;
}
}
else
{
// Special Opcodes
// A special opcode value is chosen based on the amount that needs
// to be added to the line and address registers. The maximum line
// increment for a special opcode is the value of the line_base
// field in the header, plus the value of the line_range field,
// minus 1 (line base + line range - 1). If the desired line
// increment is greater than the maximum line increment, a standard
// opcode must be used instead of a special opcode. The “address
// advance” is calculated by dividing the desired address increment
// by the minimum_instruction_length field from the header. The
// special opcode is then calculated using the following formula:
//
// opcode = (desired line increment - line_base) + (line_range * address advance) + opcode_base
//
// If the resulting opcode is greater than 255, a standard opcode
// must be used instead.
//
// To decode a special opcode, subtract the opcode_base from the
// opcode itself to give the adjusted opcode. The amount to
// increment the address register is the result of the adjusted
// opcode divided by the line_range multiplied by the
// minimum_instruction_length field from the header. That is:
//
// address increment = (adjusted opcode / line_range) * minimum_instruction_length
//
// The amount to increment the line register is the line_base plus
// the result of the adjusted opcode modulo the line_range. That is:
//
// line increment = line_base + (adjusted opcode % line_range)
uint8_t adjust_opcode = opcode - prologue->opcode_base;
dw_addr_t addr_offset = (adjust_opcode / prologue->line_range) * prologue->min_inst_length;
int32_t line_offset = prologue->line_base + (adjust_opcode % prologue->line_range);
state.line += line_offset;
state.address += addr_offset;
state.AppendRowToMatrix(*offset_ptr);
}
}
state.Finalize( *offset_ptr );
return end_offset;
}
//----------------------------------------------------------------------
// ParseStatementTableCallback
//----------------------------------------------------------------------
static void
ParseStatementTableCallback(dw_offset_t offset, const DWARFDebugLine::State& state, void* userData)
{
DWARFDebugLine::LineTable* line_table = (DWARFDebugLine::LineTable*)userData;
if (state.row == DWARFDebugLine::State::StartParsingLineTable)
{
// Just started parsing the line table, so lets keep a reference to
// the prologue using the supplied shared pointer
line_table->prologue = state.prologue;
}
else if (state.row == DWARFDebugLine::State::DoneParsingLineTable)
{
// Done parsing line table, nothing to do for the cleanup
}
else
{
// We have a new row, lets append it
line_table->AppendRow(state);
}
}
//----------------------------------------------------------------------
// ParseStatementTable
//
// Parse a line table at offset and populate the LineTable class with
// the prologue and all rows.
//----------------------------------------------------------------------
bool
DWARFDebugLine::ParseStatementTable(const DataExtractor& debug_line_data, uint32_t* offset_ptr, LineTable* line_table)
{
return ParseStatementTable(debug_line_data, offset_ptr, ParseStatementTableCallback, line_table);
}
inline bool
DWARFDebugLine::Prologue::IsValid() const
{
return SymbolFileDWARF::SupportedVersion(version);
}
//----------------------------------------------------------------------
// DWARFDebugLine::Prologue::Dump
//----------------------------------------------------------------------
void
DWARFDebugLine::Prologue::Dump(Log *log)
{
uint32_t i;
log->Printf( "Line table prologue:");
log->Printf( " total_length: 0x%8.8x", total_length);
log->Printf( " version: %u", version);
log->Printf( "prologue_length: 0x%8.8x", prologue_length);
log->Printf( "min_inst_length: %u", min_inst_length);
log->Printf( "default_is_stmt: %u", default_is_stmt);
log->Printf( " line_base: %i", line_base);
log->Printf( " line_range: %u", line_range);
log->Printf( " opcode_base: %u", opcode_base);
for (i=0; i<standard_opcode_lengths.size(); ++i)
{
log->Printf( "standard_opcode_lengths[%s] = %u", DW_LNS_value_to_name(i+1), standard_opcode_lengths[i]);
}
if (!include_directories.empty())
{
for (i=0; i<include_directories.size(); ++i)
{
log->Printf( "include_directories[%3u] = '%s'", i+1, include_directories[i].c_str());
}
}
if (!file_names.empty())
{
log->PutCString (" Dir Mod Time File Len File Name");
log->PutCString (" ---- ---------- ---------- ---------------------------");
for (i=0; i<file_names.size(); ++i)
{
const FileNameEntry& fileEntry = file_names[i];
log->Printf ("file_names[%3u] %4u 0x%8.8x 0x%8.8x %s",
i+1,
fileEntry.dir_idx,
fileEntry.mod_time,
fileEntry.length,
fileEntry.name.c_str());
}
}
}
//----------------------------------------------------------------------
// DWARFDebugLine::ParsePrologue::Append
//
// Append the contents of the prologue to the binary stream buffer
//----------------------------------------------------------------------
//void
//DWARFDebugLine::Prologue::Append(BinaryStreamBuf& buff) const
//{
// uint32_t i;
//
// buff.Append32(total_length);
// buff.Append16(version);
// buff.Append32(prologue_length);
// buff.Append8(min_inst_length);
// buff.Append8(default_is_stmt);
// buff.Append8(line_base);
// buff.Append8(line_range);
// buff.Append8(opcode_base);
//
// for (i=0; i<standard_opcode_lengths.size(); ++i)
// buff.Append8(standard_opcode_lengths[i]);
//
// for (i=0; i<include_directories.size(); ++i)
// buff.AppendCStr(include_directories[i].c_str());
// buff.Append8(0); // Terminate the include directory section with empty string
//
// for (i=0; i<file_names.size(); ++i)
// {
// buff.AppendCStr(file_names[i].name.c_str());
// buff.Append32_as_ULEB128(file_names[i].dir_idx);
// buff.Append32_as_ULEB128(file_names[i].mod_time);
// buff.Append32_as_ULEB128(file_names[i].length);
// }
// buff.Append8(0); // Terminate the file names section with empty string
//}
bool DWARFDebugLine::Prologue::GetFile(uint32_t file_idx, std::string& path, std::string& directory) const
{
uint32_t idx = file_idx - 1; // File indexes are 1 based...
if (idx < file_names.size())
{
path = file_names[idx].name;
uint32_t dir_idx = file_names[idx].dir_idx - 1;
if (dir_idx < include_directories.size())
directory = include_directories[dir_idx];
else
directory.clear();
return true;
}
return false;
}
//----------------------------------------------------------------------
// DWARFDebugLine::LineTable::Dump
//----------------------------------------------------------------------
void
DWARFDebugLine::LineTable::Dump(Log *log) const
{
if (prologue.get())
prologue->Dump (log);
if (!rows.empty())
{
log->PutCString ("Address Line Column File ISA Flags");
log->PutCString ("------------------ ------ ------ ------ --- -------------");
Row::const_iterator pos = rows.begin();
Row::const_iterator end = rows.end();
while (pos != end)
{
(*pos).Dump (log);
++pos;
}
}
}
void
DWARFDebugLine::LineTable::AppendRow(const DWARFDebugLine::Row& state)
{
rows.push_back(state);
}
//----------------------------------------------------------------------
// Compare function for the binary search in DWARFDebugLine::LineTable::LookupAddress()
//----------------------------------------------------------------------
static bool FindMatchingAddress (const DWARFDebugLine::Row& row1, const DWARFDebugLine::Row& row2)
{
return row1.address < row2.address;
}
//----------------------------------------------------------------------
// DWARFDebugLine::LineTable::LookupAddress
//----------------------------------------------------------------------
uint32_t
DWARFDebugLine::LineTable::LookupAddress(dw_addr_t address, dw_addr_t cu_high_pc) const
{
Looking at some of the test suite failures in DWARF in .o files with the debug map showed that the location lists in the .o files needed some refactoring in order to work. The case that was failing was where a function that was in the "__TEXT.__textcoal_nt" in the .o file, and in the "__TEXT.__text" section in the main executable. This made symbol lookup fail due to the way we were finding a real address in the debug map which was by finding the section that the function was in in the .o file and trying to find this in the main executable. Now the section list supports finding a linked address in a section or any child sections. After fixing this, we ran into issue that were due to DWARF and how it represents locations lists. DWARF makes a list of address ranges and expressions that go along with those address ranges. The location addresses are expressed in terms of a compile unit address + offset. This works fine as long as nothing moves around. When stuff moves around and offsets change between the remapped compile unit base address and the new function address, then we can run into trouble. To deal with this, we now store supply a location list slide amount to any location list expressions that will allow us to make the location list addresses into zero based offsets from the object that owns the location list (always a function in our case). With these fixes we can now re-link random address ranges inside the debugger for use with our DWARF + debug map, incremental linking, and more. Another issue that arose when doing the DWARF in the .o files was that GCC 4.2 emits a ".debug_aranges" that only mentions functions that are externally visible. This makes .debug_aranges useless to us and we now generate a real address range lookup table in the DWARF parser at the same time as we index the name tables (that are needed because .debug_pubnames is just as useless). llvm-gcc doesn't generate a .debug_aranges section, though this could be fixed, we aren't going to rely upon it. Renamed a bunch of "UINT_MAX" to "UINT32_MAX". llvm-svn: 113829
2010-09-14 10:20:48 +08:00
uint32_t index = UINT32_MAX;
if (!rows.empty())
{
// Use the lower_bound algorithm to perform a binary search since we know
// that our line table data is ordered by address.
DWARFDebugLine::Row row;
row.address = address;
Row::const_iterator begin_pos = rows.begin();
Row::const_iterator end_pos = rows.end();
Row::const_iterator pos = lower_bound(begin_pos, end_pos, row, FindMatchingAddress);
if (pos == end_pos)
{
if (address < cu_high_pc)
return rows.size()-1;
}
else
{
// Rely on fact that we are using a std::vector and we can do
// pointer arithmetic to find the row index (which will be one less
// that what we found since it will find the first position after
// the current address) since std::vector iterators are just
// pointers to the container type.
index = pos - begin_pos;
if (pos->address > address)
{
if (index > 0)
--index;
else
Looking at some of the test suite failures in DWARF in .o files with the debug map showed that the location lists in the .o files needed some refactoring in order to work. The case that was failing was where a function that was in the "__TEXT.__textcoal_nt" in the .o file, and in the "__TEXT.__text" section in the main executable. This made symbol lookup fail due to the way we were finding a real address in the debug map which was by finding the section that the function was in in the .o file and trying to find this in the main executable. Now the section list supports finding a linked address in a section or any child sections. After fixing this, we ran into issue that were due to DWARF and how it represents locations lists. DWARF makes a list of address ranges and expressions that go along with those address ranges. The location addresses are expressed in terms of a compile unit address + offset. This works fine as long as nothing moves around. When stuff moves around and offsets change between the remapped compile unit base address and the new function address, then we can run into trouble. To deal with this, we now store supply a location list slide amount to any location list expressions that will allow us to make the location list addresses into zero based offsets from the object that owns the location list (always a function in our case). With these fixes we can now re-link random address ranges inside the debugger for use with our DWARF + debug map, incremental linking, and more. Another issue that arose when doing the DWARF in the .o files was that GCC 4.2 emits a ".debug_aranges" that only mentions functions that are externally visible. This makes .debug_aranges useless to us and we now generate a real address range lookup table in the DWARF parser at the same time as we index the name tables (that are needed because .debug_pubnames is just as useless). llvm-gcc doesn't generate a .debug_aranges section, though this could be fixed, we aren't going to rely upon it. Renamed a bunch of "UINT_MAX" to "UINT32_MAX". llvm-svn: 113829
2010-09-14 10:20:48 +08:00
index = UINT32_MAX;
}
}
}
return index; // Failed to find address
}
//----------------------------------------------------------------------
// DWARFDebugLine::Row::Row
//----------------------------------------------------------------------
DWARFDebugLine::Row::Row(bool default_is_stmt) :
address(0),
line(1),
column(0),
file(1),
is_stmt(default_is_stmt),
basic_block(false),
end_sequence(false),
prologue_end(false),
epilogue_begin(false),
isa(0)
{
}
//----------------------------------------------------------------------
// Called after a row is appended to the matrix
//----------------------------------------------------------------------
void
DWARFDebugLine::Row::PostAppend()
{
basic_block = false;
prologue_end = false;
epilogue_begin = false;
}
//----------------------------------------------------------------------
// DWARFDebugLine::Row::Reset
//----------------------------------------------------------------------
void
DWARFDebugLine::Row::Reset(bool default_is_stmt)
{
address = 0;
line = 1;
column = 0;
file = 1;
is_stmt = default_is_stmt;
basic_block = false;
end_sequence = false;
prologue_end = false;
epilogue_begin = false;
isa = 0;
}
//----------------------------------------------------------------------
// DWARFDebugLine::Row::Dump
//----------------------------------------------------------------------
void
DWARFDebugLine::Row::Dump(Log *log) const
{
log->Printf( "0x%16.16llx %6u %6u %6u %3u %s%s%s%s%s",
address,
line,
column,
file,
isa,
is_stmt ? " is_stmt" : "",
basic_block ? " basic_block" : "",
prologue_end ? " prologue_end" : "",
epilogue_begin ? " epilogue_begin" : "",
end_sequence ? " end_sequence" : "");
}
//----------------------------------------------------------------------
// Compare function LineTable structures
//----------------------------------------------------------------------
static bool AddressLessThan (const DWARFDebugLine::Row& a, const DWARFDebugLine::Row& b)
{
return a.address < b.address;
}
// Insert a row at the correct address if the addresses can be out of
// order which can only happen when we are linking a line table that
// may have had it's contents rearranged.
void
DWARFDebugLine::Row::Insert(Row::collection& state_coll, const Row& state)
{
// If we don't have anything yet, or if the address of the last state in our
// line table is less than the current one, just append the current state
if (state_coll.empty() || AddressLessThan(state_coll.back(), state))
{
state_coll.push_back(state);
}
else
{
// Do a binary search for the correct entry
pair<Row::iterator, Row::iterator> range(equal_range(state_coll.begin(), state_coll.end(), state, AddressLessThan));
// If the addresses are equal, we can safely replace the previous entry
// with the current one if the one it is replacing is an end_sequence entry.
// We currently always place an extra end sequence when ever we exit a valid
// address range for a function in case the functions get rearranged by
// optimizations or by order specifications. These extra end sequences will
// disappear by getting replaced with valid consecutive entries within a
// compile unit if there are no gaps.
if (range.first == range.second)
{
state_coll.insert(range.first, state);
}
else
{
if ((distance(range.first, range.second) == 1) && range.first->end_sequence == true)
{
*range.first = state;
}
else
{
state_coll.insert(range.second, state);
}
}
}
}
void
DWARFDebugLine::Row::Dump(Log *log, const Row::collection& state_coll)
{
std::for_each (state_coll.begin(), state_coll.end(), bind2nd(std::mem_fun_ref(&Row::Dump),log));
}
//----------------------------------------------------------------------
// DWARFDebugLine::State::State
//----------------------------------------------------------------------
DWARFDebugLine::State::State(Prologue::shared_ptr& p, Log *l, DWARFDebugLine::State::Callback cb, void* userData) :
Row (p->default_is_stmt),
prologue (p),
log (l),
callback (cb),
callbackUserData (userData),
row (StartParsingLineTable)
{
// Call the callback with the initial row state of zero for the prologue
if (callback)
callback(0, *this, callbackUserData);
}
//----------------------------------------------------------------------
// DWARFDebugLine::State::Reset
//----------------------------------------------------------------------
void
DWARFDebugLine::State::Reset()
{
Row::Reset(prologue->default_is_stmt);
}
//----------------------------------------------------------------------
// DWARFDebugLine::State::AppendRowToMatrix
//----------------------------------------------------------------------
void
DWARFDebugLine::State::AppendRowToMatrix(dw_offset_t offset)
{
// Each time we are to add an entry into the line table matrix
// call the callback function so that someone can do something with
// the current state of the state machine (like build a line table
// or dump the line table!)
if (log)
{
if (row == 0)
{
log->PutCString ("Address Line Column File ISA Flags");
log->PutCString ("------------------ ------ ------ ------ --- -------------");
}
Dump (log);
}
++row; // Increase the row number before we call our callback for a real row
if (callback)
callback(offset, *this, callbackUserData);
PostAppend();
}
//----------------------------------------------------------------------
// DWARFDebugLine::State::Finalize
//----------------------------------------------------------------------
void
DWARFDebugLine::State::Finalize(dw_offset_t offset)
{
// Call the callback with a special row state when we are done parsing a
// line table
row = DoneParsingLineTable;
if (callback)
callback(offset, *this, callbackUserData);
}
//void
//DWARFDebugLine::AppendLineTableData
//(
// const DWARFDebugLine::Prologue* prologue,
// const DWARFDebugLine::Row::collection& state_coll,
// const uint32_t addr_size,
// BinaryStreamBuf &debug_line_data
//)
//{
// if (state_coll.empty())
// {
// // We have no entries, just make an empty line table
// debug_line_data.Append8(0);
// debug_line_data.Append8(1);
// debug_line_data.Append8(DW_LNE_end_sequence);
// }
// else
// {
// DWARFDebugLine::Row::const_iterator pos;
// Row::const_iterator end = state_coll.end();
// bool default_is_stmt = prologue->default_is_stmt;
// const DWARFDebugLine::Row reset_state(default_is_stmt);
// const DWARFDebugLine::Row* prev_state = &reset_state;
// const int32_t max_line_increment_for_special_opcode = prologue->MaxLineIncrementForSpecialOpcode();
// for (pos = state_coll.begin(); pos != end; ++pos)
// {
// const DWARFDebugLine::Row& curr_state = *pos;
// int32_t line_increment = 0;
// dw_addr_t addr_offset = curr_state.address - prev_state->address;
// dw_addr_t addr_advance = (addr_offset) / prologue->min_inst_length;
// line_increment = (int32_t)(curr_state.line - prev_state->line);
//
// // If our previous state was the reset state, then let's emit the
// // address to keep GDB's DWARF parser happy. If we don't start each
// // sequence with a DW_LNE_set_address opcode, the line table won't
// // get slid properly in GDB.
//
// if (prev_state == &reset_state)
// {
// debug_line_data.Append8(0); // Extended opcode
// debug_line_data.Append32_as_ULEB128(addr_size + 1); // Length of opcode bytes
// debug_line_data.Append8(DW_LNE_set_address);
// debug_line_data.AppendMax64(curr_state.address, addr_size);
// addr_advance = 0;
// }
//
// if (prev_state->file != curr_state.file)
// {
// debug_line_data.Append8(DW_LNS_set_file);
// debug_line_data.Append32_as_ULEB128(curr_state.file);
// }
//
// if (prev_state->column != curr_state.column)
// {
// debug_line_data.Append8(DW_LNS_set_column);
// debug_line_data.Append32_as_ULEB128(curr_state.column);
// }
//
// // Don't do anything fancy if we are at the end of a sequence
// // as we don't want to push any extra rows since the DW_LNE_end_sequence
// // will push a row itself!
// if (curr_state.end_sequence)
// {
// if (line_increment != 0)
// {
// debug_line_data.Append8(DW_LNS_advance_line);
// debug_line_data.Append32_as_SLEB128(line_increment);
// }
//
// if (addr_advance > 0)
// {
// debug_line_data.Append8(DW_LNS_advance_pc);
// debug_line_data.Append32_as_ULEB128(addr_advance);
// }
//
// // Now push the end sequence on!
// debug_line_data.Append8(0);
// debug_line_data.Append8(1);
// debug_line_data.Append8(DW_LNE_end_sequence);
//
// prev_state = &reset_state;
// }
// else
// {
// if (line_increment || addr_advance)
// {
// if (line_increment > max_line_increment_for_special_opcode)
// {
// debug_line_data.Append8(DW_LNS_advance_line);
// debug_line_data.Append32_as_SLEB128(line_increment);
// line_increment = 0;
// }
//
// uint32_t special_opcode = (line_increment >= prologue->line_base) ? ((line_increment - prologue->line_base) + (prologue->line_range * addr_advance) + prologue->opcode_base) : 256;
// if (special_opcode > 255)
// {
// // Both the address and line won't fit in one special opcode
// // check to see if just the line advance will?
// uint32_t special_opcode_line = ((line_increment >= prologue->line_base) && (line_increment != 0)) ?
// ((line_increment - prologue->line_base) + prologue->opcode_base) : 256;
//
//
// if (special_opcode_line > 255)
// {
// // Nope, the line advance won't fit by itself, check the address increment by itself
// uint32_t special_opcode_addr = addr_advance ?
// ((0 - prologue->line_base) + (prologue->line_range * addr_advance) + prologue->opcode_base) : 256;
//
// if (special_opcode_addr > 255)
// {
// // Neither the address nor the line will fit in a
// // special opcode, we must manually enter both then
// // do a DW_LNS_copy to push a row (special opcode
// // automatically imply a new row is pushed)
// if (line_increment != 0)
// {
// debug_line_data.Append8(DW_LNS_advance_line);
// debug_line_data.Append32_as_SLEB128(line_increment);
// }
//
// if (addr_advance > 0)
// {
// debug_line_data.Append8(DW_LNS_advance_pc);
// debug_line_data.Append32_as_ULEB128(addr_advance);
// }
//
// // Now push a row onto the line table manually
// debug_line_data.Append8(DW_LNS_copy);
//
// }
// else
// {
// // The address increment alone will fit into a special opcode
// // so modify our line change, then issue a special opcode
// // for the address increment and it will push a row into the
// // line table
// if (line_increment != 0)
// {
// debug_line_data.Append8(DW_LNS_advance_line);
// debug_line_data.Append32_as_SLEB128(line_increment);
// }
//
// // Advance of line and address will fit into a single byte special opcode
// // and this will also push a row onto the line table
// debug_line_data.Append8(special_opcode_addr);
// }
// }
// else
// {
// // The line change alone will fit into a special opcode
// // so modify our address increment first, then issue a
// // special opcode for the line change and it will push
// // a row into the line table
// if (addr_advance > 0)
// {
// debug_line_data.Append8(DW_LNS_advance_pc);
// debug_line_data.Append32_as_ULEB128(addr_advance);
// }
//
// // Advance of line and address will fit into a single byte special opcode
// // and this will also push a row onto the line table
// debug_line_data.Append8(special_opcode_line);
// }
// }
// else
// {
// // Advance of line and address will fit into a single byte special opcode
// // and this will also push a row onto the line table
// debug_line_data.Append8(special_opcode);
// }
// }
// prev_state = &curr_state;
// }
// }
// }
//}