def make_offsets(segname):
segea = idc.SegByBase(idc.SegByName(segname))
segend = idc.SegEnd(segea)
while segea < segend:
idc.OpOffset(segea, 0)
ptr = idc.Dword(segea)
idc.OpOffset(ptr, 0)
segea += 4
def processDataSegments(M, new_eas):
for n in xrange(idaapi.get_segm_qty()):
seg = idaapi.getnseg(n)
ea = seg.startEA
segtype = idc.GetSegmentAttr(ea, idc.SEGATTR_TYPE)
if segtype in [idc.SEG_DATA, idc.SEG_BSS]:
start = idc.SegStart(ea)
end = idc.SegEnd(ea)
addDataSegment(M, start, end, new_eas)
def define_functions():
# The function first searches for all user defined functions, reads
# the opcodes and searches for that opcodes in the rest of the file.
#
# You can extend this by adding more disassembled instructions that
# make you believe are function prologues.
#
# Obviously not any PUSH is a function start, this is only a filter
# against erroneously defined functions. So if you define a function
# that starts with other instruction (and you think there could be
# other functions that start with that instruction), just add it here.
prologues = ["STMFD", "push", "PUSH", "mov", "MOV"]
print "Finding all signatures"
#ea= 0
ea = 0x11800000
#ea = 0x10000000
opcodes = set()
if idc.SegStart(ea) == 0xFFFFFFFF:
print "Wrong segment address set!"
return
for funcea in idautils.Functions(idc.SegStart(ea), idc.SegEnd(ea)):
# Get the opcode
start_opcode = idc.Dword(funcea)
# Get the disassembled text
dis_text = idc.GetDisasm(funcea)
we_like_it = False
# Filter possible errors on manually defined functions
for prologue in prologues:
if prologue in dis_text:
we_like_it = True
# If it passes the filter, add the opcode to the search list.
if we_like_it:
opcodes.add(start_opcode)
print "# different opcodes: %x" % (len(opcodes))
while len(opcodes) > 0:
# Search for this opcode in the rest of the file
opcode_bin = opcodes.pop()
opcode_str = " ".join(
x.encode("hex") for x in struct.pack("<L", opcode_bin))
print "Searching for " + opcode_str
matches = find_all(opcode_str)
for matchea in matches:
# If the opcode is found in a non-function
if not idc.GetFunctionName(matchea):
# Try to make code and function
print "Function at " + hex(matchea)
idc.MakeCode(matchea)
idc.MakeFunction(matchea)
print "We're done!"
def nextCompressedData(ea, end_ea=None):
if not end_ea:
end_ea = idc.SegEnd(ea)
while ea < end_ea:
if mt.getLZ77CompressedSize(ea) >= 0:
return ea
ea += 1
return -1
def recover_frame_entries(seg_ea):
if seg_ea == idc.BADADDR:
return
DEBUG("Recover entries from section : {}".format(idc.SegName(seg_ea)))
ea = idc.SegStart(seg_ea)
end_ea = idc.SegEnd(seg_ea)
while ea != idc.BADADDR and ea < end_ea:
ea = format_entries(ea)
def GetROSections():
seg_addrs_readonly = []
for seg in idautils.Segments():
seg_start_ea = idc.SegStart(seg)
seg_end_ea = idc.SegEnd(seg)
flags = idc.GetSegmentAttr(seg, idc.SEGATTR_PERM)
if flags & 6: # Flags 4 means read-only
seg_addrs_readonly.append([seg_start_ea, seg_end_ea])
return seg_addrs_readonly
def get_list_of_functions(self):
'''Get all functions list.'''
seg_ea = idc.BeginEA()
functions_list = {}
for func_ea in idautils.Functions(idc.SegStart(seg_ea),
idc.SegEnd(seg_ea)):
function_name = self.maybe_demangle(idc.GetFunctionName(func_ea))
functions_list[function_name] = func_ea
return functions_list
def get_functions():
'''
enumerate the functions in the currently loaded module.
Yields:
int: address of the function.
'''
startea = idc.BeginEA()
for fva in idautils.Functions(idc.SegStart(startea), idc.SegEnd(startea)):
yield fva
def setBPs(self):
"""
Set breakpoints on all CALL and RET instructions in all of the executable sections.
"""
for seg_ea in idautils.Segments():
for head in idautils.Heads(seg_ea, idc.SegEnd(seg_ea)):
if idc.isCode(idc.GetFlags(head)):
# Add BP if instruction is a CALL
if is_call(head):
self.addBP(head)
def apply_everywhere():
hr_remove()
for segea in idautils.Segments():
for funcea in idautils.Functions(segea, idc.SegEnd(segea)):
print("Handling %s" % idc.GetFunctionName(funcea))
try:
cfunc = ida_hexrays.decompile(funcea)
apply_on_fn(cfunc)
except ida_hexrays.DecompilationFailure as e:
print("Failed to decompile!")
hr_install()
def analyse_subroutines():
"""Goes through all the subroutines that IDA's initial auto analysis
discovers."""
log.info("Analysing subroutines")
exported_eas = find_exported_eas()
log.info("IDA identified {} exported functions".format(len(exported_eas)))
subs = set()
sub_eas = set() # Work list.
sub_eas.update(exported_eas)
# Get all subroutines that IDA recognised.
for seg_ea in idautils.Segments():
min_ea, max_ea = idc.SegStart(seg_ea), idc.SegEnd(seg_ea)
sub_eas.update(idautils.Functions(min_ea, max_ea))
log.info("IDA identified {} functions".format(len(sub_eas)))
bad_sub_eas = set()
for sub_ea in sub_eas:
if has_segment_type(sub_ea, idc.SEG_CODE):
sub = program.get_subroutine(sub_ea) # Mark `ea` as a subroutine.
# Iteratively analyse the blocks in subroutines. This may discover new
# subroutines because our block analysis can be more aggressive than what
# IDA finds.
while len(sub_eas):
sub_ea = sub_eas.pop()
if not has_segment_type(sub_ea, idc.SEG_CODE):
log.warning(
"Not analysing subroutine at non-code address {:08x}".format(
sub_ea))
continue
sub = program.get_subroutine(sub_ea)
if sub in subs:
log.debug("Skipping {:08x}; already analysed.".format(sub_ea))
continue
subs.add(sub)
if idc.hasName(sub_ea):
sub.name = idc.GetFunctionName(sub_ea)
# Mark this subroutine as exported.
if sub_ea in exported_eas:
sub.visibility = program.Subroutine.VISIBILITY_EXPORTED
analyse_subroutine(sub)
sub_eas.update(get_called_subroutines(sub))
return subs
def main():
objc_data_seg = idc.SegByBase(idc.SegByName('__objc_data'))
if objc_data_seg == idc.BADADDR:
print 'Cannot locate objc_data segment'
return
ea = objc_data_seg
while ea < idc.SegEnd(objc_data_seg):
objc_class = ObjcClass(ea)
objc_class.dump()
ea = ea + 0x14
def get_sections():
"""Get section names and start/end addrs from IDA database
"""
sections = {}
for ea in idautils.Segments():
curr = {}
curr["start"] = idc.SegStart(ea)
curr["end"] = idc.SegEnd(ea)
sections[idc.SegName(ea)] = curr
return sections
def promptForRange(self):
#check if a range has already been selected - if so skip prompt
start = idc.SelStart()
if start != idc.BADADDR:
self.logger.info('Processing range 0x%08x - 0x%08x', self.params.startAddr, self.params.endAddr)
self.params.startAddr = start
self.params.endAddr = idc.SelEnd()
else:
self.logger.info('Processing current segment only')
self.params.startAddr = idc.SegStart(idc.here())
self.params.endAddr = idc.SegEnd(idc.here())
def getBinary(self):
result = b""
segment_starts = [ea for ea in idautils.Segments()]
offsets = []
start_len = 0
for start in segment_starts:
end = idc.SegEnd(start)
result += idc.get_bytes(start, end - start)
offsets.append((start, start_len, len(result)))
start_len = len(result)
return result
def update_mapping(self):
pass
self.fun_mapping = {
idc.GetFunctionName(x):
(idaapi.get_func(x).startEA, idaapi.get_func(x).endEA - 1)
for x in idautils.Functions()
}
self.seg_mapping = {
idc.SegName(x): (idc.SegStart(x), idc.SegEnd(x))
for x in idautils.Segments()
}
def recordText(self):
for seg_ea in idautils.Segments():
print('seg: %s' % idc.SegName(seg_ea))
if idc.SegName(seg_ea) == '.text':
start = idc.SegStart(seg_ea)
end = idc.SegEnd(seg_ea)
print('text at 0x%x - 0x%x' % (start, end))
gdbProt.Evalx(
'SendGDBMonitor("@cgc.recordText(0x%x, 0x%x)");' %
(start, end))
break
def _process_offsets_section(segstart, next_offset):
"""Process all the offsets in a __got section."""
for offset, ea in idau.ReadWords(segstart,
idc.SegEnd(segstart),
addresses=True):
if not offset_name_target(idau.get_ea_name(ea)):
# This is not a previously named offset.
if idau.is_mapped(offset, value=False):
_process_offset(offset, ea, next_offset)
else:
_log(-1, 'Offset {:#x} at address {:#x} is unmapped', offset,
ea)
def __init__(self, fmt='[+] srch (IDB/binary searching utils)'):
"""
This module is responsible for printing disassemblies and necessary compoents
of disassemblies
"""
super(srch, self).__init__(fmt)
self.registerCommand(self, self.nextarm, "nextarm", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextascii, "nextascii", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextfakeinst, "nextfakeinst", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextname, "nextname", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextknown, "nextknown", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextbin, "nextbin", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextred, "nextred", "<search_ea> [ui=True]")
self.registerCommand(self, self.nextimmref, "nextimmref", "<search_ea> [ui=True]")
# figure out the very last ea reachable
self.end_ea = 0
for seg in idautils.Segments():
if idc.SegEnd(seg) > self.end_ea:
self.end_ea = idc.SegEnd(seg)
def get_instruction_dictionary(ea):
insts = dict()
for func in idautils.Functions(idc.SegStart(ea), idc.SegEnd(ea)):
f_end = FindFuncEnd(func)
for head in Heads(func, f_end):
if not isCode(GetFlags(head)):
continue
mnem = GetMnem(head)
if mnem not in insts:
insts[mnem] = list()
insts[mnem].append(head)
return insts
def make_offsets(segname):
'''
change the segment's data value into offset by class name
'''
segea = idc.SegByBase(idc.SegByName(segname))
segend = idc.SegEnd(segea)
while segea < segend:
idc.OpOffset(segea, 0)
ptr = idc.Dword(segea)
idc.OpOffset(ptr, 0)
segea += 4
def _get_memory(self):
result = ""
segment_starts = [ea for ea in idautils.Segments()]
offsets = []
start_len = 0
for start in segment_starts:
end = idc.SegEnd(start)
for ea in lrange(start, end):
result += chr(idc.Byte(ea))
offsets.append((start, start_len, len(result)))
start_len = len(result)
return result, offsets
def _process_const_section_for_vtables(segstart, metaclass_info, found_vtable):
"""Process a __const section to search for virtual method tables."""
segend = idc.SegEnd(segstart)
addr = segstart
while addr < segend:
possible, length = vtable.vtable_length(addr, segend, scan=True)
if possible:
metaclass = _get_vtable_metaclass(addr, metaclass_info)
if metaclass:
_log(4, 'Vtable at address {:#x} has metaclass {:#x}', addr, metaclass)
found_vtable(metaclass, addr, length)
addr += length * idau.WORD_SIZE
def getSegsInfo():
'''
Returns a list of all segments in the form: (name, segstart, segend)
'''
segments = list(idautils.Segments())
res = []
for s in segments:
res.append((idc.SegName(s), idc.SegStart(s), idc.SegEnd(s)))
return res
def processCode(self):
if (self.params.startAddr==idc.BADADDR) and (self.params.endAddr==idc.BADADDR):
self.logger.info('Processing current segment only')
#self.processAllSegments()
self.params.startAddr = idc.SegStart(idc.here())
self.params.endAddr = idc.SegEnd(idc.here())
else:
self.logger.info('Processing range 0x%08x - 0x%08x', self.params.startAddr, self.params.endAddr)
if self.params.searchDwordArray:
self.lookForDwordArray(self.params.startAddr, self.params.endAddr)
if self.params.searchPushArgs:
self.lookForOpArgs(self.params.startAddr, self.params.endAddr)
def get_import_module_name(address):
global module_names
global idata_seg_start
global idata_seg_end
segment_eas = list(idautils.Segments())
# This hasn't been initialized yet...
#
if module_names is None:
module_names = list()
for idata_seg_start in segment_eas:
print "Going through segment %08X" % idata_seg_start
segment = idaapi.getseg(idata_seg_start)
if segment.type != idaapi.SEG_XTRN:
continue
print "Found idata segment"
idata_seg_end = idc.SegEnd(idata_seg_start)
parse = re.compile('.*Imports\s+from\s+([\w\d]+\.[\w\d]+).*',
re.IGNORECASE)
# save the address/module name combinations we discover
#
modules = list()
# Scan the .idata segment looking for the imports from
# string and get the address ranges where it applies
#
for head in idautils.Heads(idata_seg_start, idata_seg_end):
for line_id in range(100):
line = idc.LineA(head, line_id)
if line and 'imports from' in line.lower():
res = parse.match(line)
if res:
print 'Found import line [%s][%s]' % (line,
res.group(1))
modules.append((head, res.group(1).lower()))
modules.append((idata_seg_end, None))
for idx in range(len(modules) - 1):
mod = modules[idx]
module_names.append(((mod[0], modules[idx + 1][0]), mod[1]))
for addr_range, module_name in module_names:
if addr_range[0] <= address < addr_range[1]:
return module_name
return None
def analyse_data(pointer_size):
"""Go through the data sections and look for possible tables of
code pointers."""
log.info("Analysing the data section for simple code refs")
for seg_ea in idautils.Segments():
seg_type = idc.GetSegmentAttr(seg_ea, idc.SEGATTR_TYPE)
if seg_type != idc.SEG_DATA:
continue
min_ea, max_ea = idc.SegStart(seg_ea), idc.SegEnd(seg_ea)
if 8 == pointer_size:
scan_data_for_code_refs(min_ea, max_ea, idc.Qword, 8)
scan_data_for_code_refs(min_ea, max_ea, idc.Dword, 4)
def processAllSegments(self):
for seg in idautils.Segments():
if using_ida7api:
segStart = idc.get_segm_start(seg)
segEnd = idc.get_segm_end(seg)
else:
segStart = idc.SegStart(seg)
segEnd = idc.SegEnd(seg)
if self.params.searchPushArgs:
self.lookForOpArgs(segStart, segEnd)
if self.params.searchDwordArray:
self.lookForDwordArray(segStart, segEnd)
def iter_lines():
"""
Iterate through all line addresses in the IDB
Yields addresses of all lines.
"""
for ea in idautils.Segments():
seg_start = idc.SegStart(ea)
seg_end = idc.SegEnd(ea)
cur_addr = seg_start
while (cur_addr < seg_end) and (cur_addr != idaapi.BADADDR):
yield cur_addr
cur_addr = idc.NextHead(cur_addr)
def walk_datas():
for seg_start in idautils.Segments():
seg_end = idc.SegEnd(seg_start)
for ea in ya.get_all_items(seg_start, seg_end):
flags = idc.GetFlags(ea)
func = idaapi.get_func(ea)
if idaapi.isFunc(flags) or (func and idc.isCode(flags)):
# function
continue
if not func and idc.isCode(flags):
# code
continue
yield ea
