esp-idf/tools/idf_size.py

#!/usr/bin/env python
#
# esp-idf alternative to "size" to print ELF file sizes, also analyzes
# the linker map file to dump higher resolution details.
#
# Includes information which is not shown in "xtensa-esp32-elf-size",
# or easy to parse from "xtensa-esp32-elf-objdump" or raw map files.
#
# SPDX-FileCopyrightText: 2017-2022 Espressif Systems (Shanghai) CO LTD
# SPDX-License-Identifier: Apache-2.0
#
from __future__ import division, print_function, unicode_literals

import argparse
import collections
import json
import os.path
import re
import sys

from future.utils import iteritems

try:
    from typing import Any, Callable, Collection, Dict, Iterable, List, Optional, TextIO, Tuple, Union
    Section = Dict[str, Union[str, int]]
    SectionDict = Dict[str, Section]
except ImportError:
    pass

try:
    basestring
except NameError:
    basestring = str


GLOBAL_JSON_INDENT = 4
GLOBAL_JSON_SEPARATORS = (',', ': ')


class MemRegions(object):
    # Regions determined by the chip target.

    # DIRAM is not added here. The DIRAM is indicated by the `secondary_addr` of each MemRegDef
    (DRAM_ID, IRAM_ID, CACHE_D_ID, CACHE_I_ID, RTC_FAST_D_ID, RTC_FAST_I_ID, RTC_SLOW_D_ID) = range(7)

    # The order of variables in the tuple is the same as in the soc_memory_layout.c files
    MemRegDef = collections.namedtuple('MemRegDef', ['primary_addr', 'length', 'type', 'secondary_addr'])

    class Region(object):
        # Helper class to store region information
        def __init__(self, start, length, region, section=None):
            # type: (MemRegions.Region, int, int, MemRegions.MemRegDef, Optional[str]) -> None
            self.start = start
            self.len = length
            self.region = region
            self.section = section

    @staticmethod
    def get_mem_regions(target):  # type: (str) -> List
        # The target specific memory structure is deduced from soc_memory_types defined in
        # $IDF_PATH/components/soc/**/soc_memory_layout.c files.

        MemRegDef = MemRegions.MemRegDef

        # Consecutive MemRegDefs of the same type are joined into one MemRegDef
        if target == 'esp32':
            return sorted([
                MemRegDef(0x3FFAE000, 17 * 0x2000 + 4 * 0x8000 + 4 * 0x4000, MemRegions.DRAM_ID, 0),
                MemRegDef(0x40070000, 2 * 0x8000 + 16 * 0x2000, MemRegions.IRAM_ID, 0),
                MemRegDef(0x400C2000, 0xB3E000, MemRegions.CACHE_I_ID, 0),
                MemRegDef(0x3F400000, 0x400000, MemRegions.CACHE_D_ID, 0),
                MemRegDef(0x3F800000, 0x400000, MemRegions.CACHE_D_ID, 0),
                MemRegDef(0x3FF80000, 0x2000, MemRegions.RTC_FAST_D_ID, 0x400C0000),
                MemRegDef(0x50000000, 0x2000, MemRegions.RTC_SLOW_D_ID, 0),
            ])
        elif target == 'esp32s2':
            return sorted([
                MemRegDef(0x3FFB2000, 3 * 0x2000 + 18 * 0x4000, MemRegions.DRAM_ID, 0x40022000),
                MemRegDef(0x3F000000, 0x400000, MemRegions.CACHE_I_ID, 0),
                MemRegDef(0x3F500000, 0xA80000, MemRegions.CACHE_D_ID, 0),
                MemRegDef(0x40080000, 0x780000, MemRegions.CACHE_I_ID, 0),
                MemRegDef(0x40070000, 0x2000, MemRegions.RTC_FAST_D_ID, 0x3FF9E000),
                MemRegDef(0x50000000, 0x2000, MemRegions.RTC_SLOW_D_ID, 0),
            ])
        else:
            raise RuntimeError('Target not detected.')

    def __init__(self, target):  # type: (MemRegions, str) -> None
        self.chip_mem_regions = self.get_mem_regions(target)
        if not self.chip_mem_regions:
            raise RuntimeError('Target {} is not implemented in idf_size'.format(target))

    def _get_first_region(self, start, length):
        # type: (int, int) -> Tuple[MemRegions.MemRegDef, int]
        for region in self.chip_mem_regions:  # type: ignore
            if region.primary_addr <= start < region.primary_addr + region.length:
                return (region, min(length, region.primary_addr + region.length - start))

            if (region.secondary_addr and region.secondary_addr <= start < region.secondary_addr + region.length):
                return (region, min(length, region.secondary_addr + region.length - start))

        raise RuntimeError('Given section not found in any memory region. '
                           'Check whether the LD file is compatible with the definitions in get_mem_regions in idf_size.py')

    def _get_regions(self, start, length, name=None):  # type: (int, int, Optional[str]) -> List
        ret = []
        while length > 0:
            (region, cur_len) = self._get_first_region(start, length)
            ret.append(MemRegions.Region(start, cur_len, region, name))
            length -= cur_len
            start += cur_len

        return ret

    def fit_segments_into_regions(self, segments):  # type: (MemRegions, Dict) -> List
        region_list = []
        for segment in segments.values():
            sorted_segments = self._get_regions(segment['origin'], segment['length'])
            region_list.extend(sorted_segments)

        return region_list

    def fit_sections_into_regions(self, sections):  # type: (MemRegions, Dict) -> List
        region_list = []

        for section in sections.values():
            sorted_sections = self._get_regions(section['address'], section['size'], section['name'])
            region_list.extend(sorted_sections)

        return region_list


class LinkingSections(object):

    _section_type_dict = {k: re.compile(v) for k, v in {
        'text': r'.*\.text',
        'data': r'.*\.data',
        'bss': r'.*\.bss',
        'rodata': r'.*\.rodata',
        'noinit': r'.*noinit',
        'vectors': r'.*\.vectors',
        'flash': r'.*flash.*',
    }.items()}

    @staticmethod
    def in_section(section, section_name_or_list):  # type: (str, Union[str, Iterable]) -> bool
        if isinstance(section_name_or_list, basestring):
            section_name_or_list = [section_name_or_list]

        for section_name in section_name_or_list:
            if LinkingSections._section_type_dict[section_name].match(section):
                return True
        return False

    @staticmethod
    def filter_sections(sections):  # type: (Dict) -> Dict
        return {k: v for k, v in sections.items()
                if LinkingSections.in_section(k, LinkingSections._section_type_dict.keys())}

    @staticmethod
    def get_display_name_order(section_name_list):  # type: (List[str]) -> Tuple[List[str], List[str]]
        '''
        Return two lists, in the suggested display order.
        First list is the reordered section_name_list, second list is the suggested display name, corresponding to the first list
        '''

        def get_name_score(name):  # type: (str) -> int
            score_dict = {
                '.dram': 30,
                '.iram': 20,
                '.flash': 10,
                'ram_st_total': -10,
                'flash_total': -20,
                '.data': 6,
                '.bss': 5,
                '.text': 4,
                '.rodata': 3,
                '.vectors': 2,
                '.noinit': 1,
                '.other': -1,
            }
            return sum([score if section in name else 0
                        for section, score in score_dict.items()])

        score_list = [get_name_score(section) for section in section_name_list]
        ordered_name_list = sorted(section_name_list, key=lambda x: score_list[section_name_list.index(x)], reverse=True)
        display_name_list = ordered_name_list[:]

        memory_name = ''
        display_name_list = sorted(display_name_list)
        ordered_name_list = sorted(ordered_name_list)
        ordered_name_list = check_is_dict_sort(ordered_name_list)
        for i, section in enumerate(ordered_name_list):
            if memory_name and section.startswith(memory_name):
                # If the section has same memory type with the previous one, use shorter name
                display_name_list[i] = section.replace(memory_name, '& ')
                continue

            memory_name = ''
            split_name = section.split('.')
            if len(split_name) > 1:
                # If the section has a memory type, update the type and try to display the type properly
                assert len(split_name) == 3 and split_name[0] == '', 'Unexpected section name'
                memory_name = '.iram' if 'iram' in split_name[1] else\
                              '.dram' if 'dram' in split_name[1] else\
                              '.flash' if 'flash' in split_name[1] else\
                              '.' + split_name[1]
                display_name_list[i] = 'DRAM .' + split_name[2] if 'dram' in split_name[1] else\
                                       'IRAM' + split_name[1].replace('iram', '') + ' .' + split_name[2] if 'iram' in split_name[1] else\
                                       'Flash .' + split_name[2] if 'flash' in split_name[1] else\
                                       section
                continue

            # Otherwise use its original name
            display_name_list[i] = section

        return ordered_name_list, display_name_list


def scan_to_header(f, header_line):  # type: (Iterable, str) -> None
    """ Scan forward in a file until you reach 'header_line', then return """
    for line in f:
        if line.strip() == header_line:
            return
    raise RuntimeError("Didn't find line '%s' in file" % header_line)


def format_json(json_object):  # type: (Dict) -> str
    return json.dumps(json_object,
                      allow_nan=True,
                      indent=GLOBAL_JSON_INDENT,
                      separators=GLOBAL_JSON_SEPARATORS) + os.linesep


def load_map_data(map_file):  # type: (TextIO) -> Tuple[str, Dict, Dict]
    segments = load_segments(map_file)
    detected_chip = detect_target_chip(map_file)
    sections = load_sections(map_file)

    # Exclude the dummy and .text_end section, which usually means shared region among I/D buses
    for key in list(sections.keys()):
        if key.endswith(('dummy', '.text_end')):
            sections.pop(key)
    return detected_chip, segments, sections


def load_segments(map_file):  # type: (TextIO) -> Dict
    """ Memory Configuration section is the total size of each segment """
    result = {}  # type: Dict[Any, Dict]
    scan_to_header(map_file, 'Memory Configuration')
    RE_MEMORY_SECTION = re.compile(r'(?P<name>[^ ]+) +0x(?P<origin>[\da-f]+) +0x(?P<length>[\da-f]+)')

    for line in map_file:
        m = RE_MEMORY_SECTION.match(line)
        if m is None:
            if len(result) == 0:
                continue  # whitespace or a header, before the content we want
            else:
                return result  # we're at the end of the Memory Configuration
        segment = {
            'name': m.group('name'),
            'origin': int(m.group('origin'), 16),
            'length': int(m.group('length'), 16),
        }
        if segment['name'] != '*default*':
            result[segment['name']] = segment
    raise RuntimeError('End of file while scanning memory configuration?')


def detect_target_chip(map_file):  # type: (Iterable) -> str
    ''' Detect target chip based on the target archive name in the linker script part of the MAP file '''
    scan_to_header(map_file, 'Linker script and memory map')

    RE_TARGET = re.compile(r'project_elf_src_(.*)\.c.obj')
    # For back-compatible with make
    RE_TARGET_MAKE = re.compile(r'^LOAD .*?/xtensa-([^-]+)-elf/')

    for line in map_file:
        m = RE_TARGET.search(line)
        if m:
            return m.group(1)

        m = RE_TARGET_MAKE.search(line)
        if m:
            return m.group(1)

        line = line.strip()
        # There could be empty line(s) between the "Linker script and memory map" header and "LOAD lines". Therefore,
        # line stripping and length is checked as well. The "LOAD lines" are between START GROUP and END GROUP for
        # older MAP files.
        if not line.startswith(('LOAD', 'START GROUP', 'END GROUP')) and len(line) > 0:
            # This break is a failsafe to not process anything load_sections() might want to analyze.
            break

    raise RuntimeError('Target not detected')


def load_sections(map_file):  # type: (TextIO) -> Dict
    """ Load section size information from the MAP file.

    Returns a dict of 'sections', where each key is a section name and the value
    is a dict with details about this section, including a "sources" key which holds a list of source file line
    information for each symbol linked into the section.

    There are two kinds of lines:
        - symbol_only: [optional space]<sym_name>
        - full line: [optional space][optional sym_name] <address> <size> [optional file_info]
          If <sym_name> doesn't exist, ues the symbol name from the symbol_only line above
          If the line is the starting of a section, the <file> should be empty, otherwise if the line is for a source
          line, the <file> must exist, or the <sym_name> should be is no *fill*. This rule is used to tell sections from
          source lines.
    """

    # Check for lines which only contain the sym name (and rest is on following lines)
    RE_SYMBOL_ONLY_LINE = re.compile(r'^\s*(?P<sym_name>\S*)$')

    # Fast check to see if line is a potential source line before running the slower full regex against it
    RE_PRE_FILTER = re.compile(r'.*0x[\da-f]+\s*0x[\da-f]+.*')

    # source file line, ie
    # 0x0000000040080400       0xa4 /home/gus/esp/32/idf/examples/get-started/hello_world/build/esp32/libesp32.a(cpu_start.o)
    # cmake build system links some object files directly, not part of any archive, so make that part optional
    #  .xtensa.info   0x0000000000000000       0x38 CMakeFiles/hello-world.elf.dir/project_elf_src.c.obj
    #  *fill*         0x00000000400e2967        0x1
    RE_FULL_LINE = re.compile(r'\s*(?P<sym_name>\S*) +0x(?P<address>[\da-f]+) +0x(?P<size>[\da-f]+)\s*(?P<file>.*)$')

    # Extract archive and object_file from the file_info field
    RE_FILE = re.compile(r'((?P<archive>[^ ]+\.a)?\(?(?P<object_file>[^ ]+\.(o|obj))\)?)')

    def dump_src_line(src):  # type: (Dict) -> str
        return '%s(%s) addr: 0x%08x, size: 0x%x+%d' % (src['sym_name'], src['file'], src['address'], src['size'], src['fill'])

    sections = {}  # type: Dict[Any, Dict]
    section = {}  # type: Dict[str, Any]
    sym_backup = ''
    for line in map_file:
        if line.strip() == 'Cross Reference Table':
            # stop processing lines because we are at the next section in the map file
            break

        m = RE_SYMBOL_ONLY_LINE.match(line)
        if m:
            # In some cases the section name appears on the previous line, back it up in here
            sym_backup = m.group('sym_name')
            continue

        if not RE_PRE_FILTER.match(line):
            # line does not match our quick check, so skip to next line
            continue

        m = RE_FULL_LINE.match(line)
        if not m:
            assert not sym_backup, 'Symbol only line must be followed by a line with address and size'
            continue

        name = m.group('sym_name') if m.group('sym_name') else sym_backup
        sym_backup = ''

        is_section = not m.group('file') and name != '*fill*'
        if is_section:
            # section

            section = {
                'name': name,
                'address': int(m.group('address'), 16),
                'size': int(m.group('size'), 16),
                'sources': [],
            }
            sections[name] = section

        else:
            # symbol
            if not section:
                continue

            # There are some source lines in rodata section doesn't actually take any space, but have size
            # Make size of those sections zero
            srcs = section['sources']  # type: List[Dict]
            if srcs:
                last_src = srcs[-1]
                if last_src['size'] > 0 and last_src['address'] == int(m.group('address'), 16):
                    if '.comment' != section['name'] and '.debug_str' != section['name'] and\
                            'rodata' not in last_src['sym_name']:

                        raise RuntimeError('Due to overlap with following lines, size of the line set to 0:\n    %s' % dump_src_line(last_src))

                    last_src['size'] = 0

            # Count the padding size into the last valid (size > 0) source in the section
            if name == '*fill*':
                for src in reversed(srcs):
                    if src['size'] > 0:
                        src['fill'] += int(m.group('size'), 16)
                        break
                continue

            # Extract archive and file information
            n = RE_FILE.match(m.group('file'))
            assert n

            archive = n.group('archive')
            if archive is None:
                # optional named group "archive" was not matched, so assign a value to it
                archive = '(exe)'

            file = n.group('object_file')

            assert name
            source = {
                'size': int(m.group('size'), 16),
                'address': int(m.group('address'), 16),
                'archive': os.path.basename(archive),
                'object_file': os.path.basename(file),
                'sym_name': name,
                'fill': 0,  # padding size ofter the source
            }
            source['file'] = '%s:%s' % (source['archive'], source['object_file'])

            section['sources'].append(source)  # type: ignore

    # Validate the map file
    for section in sections.values():
        src_curr = {}  # type: Dict[str, Any]
        for src in section['sources']:
            if src['size'] == 0:
                continue

            expected_addr = src_curr['address'] + src_curr['size'] + src_curr['fill'] if src_curr else section['sources'][0]['address']
            if src['address'] != expected_addr:
                print('Warning: source line overlap:')
                print('    ' + dump_src_line(src_curr))
                print('    ' + dump_src_line(src))

            src_curr = src

    return sections


def check_target(target, map_file):  # type: (str, TextIO) -> None
    if target is None:
        raise RuntimeError('The target chip cannot be detected for {}. '
                           'Please report the issue.'.format(map_file.name))


def main():  # type: () -> None
    parser = argparse.ArgumentParser(description='idf_size - a tool to print size information from an IDF MAP file')

    parser.add_argument(
        '--json',
        help='Output results as JSON',
        action='store_true')

    parser.add_argument(
        'map_file', help='MAP file produced by linker',
        type=argparse.FileType('r'))

    parser.add_argument(
        '--archives', help='Print per-archive sizes', action='store_true')

    parser.add_argument(
        '--archive_details', help='Print detailed symbols per archive')

    parser.add_argument(
        '--files', help='Print per-file sizes', action='store_true')

    parser.add_argument(
        '--target', help='Set target chip', default=None)

    parser.add_argument(
        '--diff', help='Show the differences in comparison with another MAP file',
        metavar='ANOTHER_MAP_FILE',
        default=None,
        dest='another_map_file')

    parser.add_argument(
        '-o',
        '--output-file',
        type=argparse.FileType('w'),
        default=sys.stdout,
        help='Print output to the specified file instead of stdout')

    args = parser.parse_args()

    detected_target, segments, sections = load_map_data(args.map_file)
    args.map_file.close()
    check_target(detected_target, args.map_file)

    if args.another_map_file:
        with open(args.another_map_file, 'r') as f:
            detected_target_diff, segments_diff, sections_diff = load_map_data(f)
            check_target(detected_target_diff, f)
            if detected_target_diff != detected_target:
                print('WARNING: The target of the reference and other MAP files is {} and {}, respectively.'
                      ''.format(detected_target, detected_target_diff))
    else:
        segments_diff, sections_diff, detected_target_diff = {}, {}, ''

    if args.target is not None:
        if args.target != detected_target or (detected_target_diff and args.target != detected_target_diff):
            print('WARNING: The detected chip target overwritten to {} by command line argument!'.format(args.target))

        detected_target = args.target
        detected_target_diff = args.target

    output = ''

    if not args.json or not (args.archives or args.files or args.archive_details):
        output += get_summary(args.map_file.name, segments, sections, detected_target,
                              args.json,
                              args.another_map_file, segments_diff, sections_diff, detected_target_diff)

    if args.archives:
        output += get_detailed_sizes(sections, 'archive', 'Archive File', args.json, sections_diff)
    if args.files:
        output += get_detailed_sizes(sections, 'file', 'Object File', args.json, sections_diff)

    if args.archive_details:
        output += get_archive_symbols(sections, args.archive_details, args.json, sections_diff)

    args.output_file.write(output)
    args.output_file.close()


class StructureForSummary(object):
    # this is from main branch
    # used_dram_data, used_dram_bss, used_dram_other, used_dram, dram_total, dram_remain = (0, ) * 6
    used_dram_data, used_dram_bss, used_dram_rodata, used_dram_other, used_dram, dram_total, dram_remain = (0, ) * 7

    used_dram_ratio = 0.
    used_iram_vectors, used_iram_text, used_iram_other, used_iram, iram_total, iram_remain = (0, ) * 6
    used_iram_ratio = 0.
    used_diram_data, used_diram_bss, used_diram_text, used_diram_vectors, used_diram_rodata, used_diram_other, diram_total, used_diram, diram_remain = (0, ) * 9
    used_diram_ratio = 0.
    used_flash_text, used_flash_rodata, used_flash_other, used_flash, total_size = (0, ) * 5

    def __sub__(self, rhs):  # type: (StructureForSummary) -> StructureForSummary
        assert isinstance(rhs, StructureForSummary)
        ret = self
        for key in StructureForSummary.get_required_items():
            setattr(ret, key, getattr(self, key) - getattr(rhs, key))

        return ret

    @classmethod
    def get_required_items(cls):  # type: (Any) -> List
        whole_list = list(filter(lambda x: not (x.startswith('__') or x.endswith('__') or callable(getattr(cls, x))), dir(cls)))
        return whole_list

    @staticmethod
    def get(segments, sections):  # type: (List, List) -> StructureForSummary

        def get_size(sections):  # type: (Iterable) -> int
            return sum([x.len for x in sections])

        def in_diram(x):  # type: (MemRegions.Region) -> bool
            return x.region.type in (MemRegions.DRAM_ID, MemRegions.IRAM_ID) and x.region.secondary_addr > 0

        def in_dram(x):  # type: (MemRegions.Region) -> bool
            return x.region.type == MemRegions.DRAM_ID and x.region.secondary_addr == 0  # type: ignore

        def in_iram(x):  # type: (MemRegions.Region) -> bool
            return x.region.type == MemRegions.IRAM_ID and x.region.secondary_addr == 0  # type: ignore

        r = StructureForSummary()

        diram_filter = filter(in_diram, segments)
        r.diram_total = int(get_size(diram_filter) / 2)

        dram_filter = filter(in_dram, segments)
        r.dram_total = get_size(dram_filter)
        iram_filter = filter(in_iram, segments)
        r.iram_total = get_size(iram_filter)

        def filter_in_section(sections, section_to_check):  # type: (Iterable[MemRegions.Region], str) -> List[MemRegions.Region]
            return list(filter(lambda x: LinkingSections.in_section(x.section, section_to_check), sections))  # type: ignore

        dram_sections = list(filter(in_dram, sections))
        iram_sections = list(filter(in_iram, sections))
        diram_sections = list(filter(in_diram, sections))
        flash_sections = filter_in_section(sections, 'flash')

        dram_data_list = filter_in_section(dram_sections, 'data')
        dram_bss_list = filter_in_section(dram_sections, 'bss')
        dram_rodata_list = filter_in_section(dram_sections, 'rodata')
        dram_other_list = [x for x in dram_sections if x not in dram_data_list + dram_bss_list + dram_rodata_list]

        iram_vectors_list = filter_in_section(iram_sections, 'vectors')
        iram_text_list = filter_in_section(iram_sections, 'text')
        iram_other_list = [x for x in iram_sections if x not in iram_vectors_list + iram_text_list]

        diram_vectors_list = filter_in_section(diram_sections, 'vectors')
        diram_data_list = filter_in_section(diram_sections, 'data')
        diram_bss_list = filter_in_section(diram_sections, 'bss')
        diram_text_list = filter_in_section(diram_sections, 'text')
        diram_rodata_list = filter_in_section(diram_sections, 'rodata')
        diram_other_list = [x for x in diram_sections if x not in diram_data_list + diram_bss_list + diram_text_list + diram_vectors_list + diram_rodata_list]

        flash_text_list = filter_in_section(flash_sections, 'text')
        flash_rodata_list = filter_in_section(flash_sections, 'rodata')
        flash_other_list = [x for x in flash_sections if x not in flash_text_list + flash_rodata_list]

        r.used_dram_data = get_size(dram_data_list)
        r.used_dram_bss = get_size(dram_bss_list)
        r.used_dram_rodata = get_size(dram_rodata_list)
        r.used_dram_other = get_size(dram_other_list)
        r.used_dram = r.used_dram_data + r.used_dram_bss + r.used_dram_other + r.used_dram_rodata
        try:
            r.used_dram_ratio = r.used_dram / r.dram_total
        except ZeroDivisionError:
            r.used_dram_ratio = float('nan') if r.used_dram != 0 else 0
        r.dram_remain = r.dram_total - r.used_dram

        r.used_iram_vectors = get_size((iram_vectors_list))
        r.used_iram_text = get_size((iram_text_list))
        r.used_iram_other = get_size((iram_other_list))
        r.used_iram = r.used_iram_vectors + r.used_iram_text + r.used_iram_other
        try:
            r.used_iram_ratio = r.used_iram / r.iram_total
        except ZeroDivisionError:
            r.used_iram_ratio = float('nan') if r.used_iram != 0 else 0
        r.iram_remain = r.iram_total - r.used_iram

        r.used_diram_data = get_size(diram_data_list)
        r.used_diram_bss = get_size(diram_bss_list)
        r.used_diram_text = get_size(diram_text_list)
        r.used_diram_vectors = get_size(diram_vectors_list)
        r.used_diram_rodata = get_size(diram_rodata_list)
        r.used_diram_other = get_size(diram_other_list)
        r.used_diram = r.used_diram_data + r.used_diram_bss + r.used_diram_text + r.used_diram_vectors + r.used_diram_other + r.used_diram_rodata
        try:
            r.used_diram_ratio = r.used_diram / r.diram_total
        except ZeroDivisionError:
            r.used_diram_ratio = float('nan') if r.used_diram != 0 else 0
        r.diram_remain = r.diram_total - r.used_diram

        r.used_flash_text = get_size(flash_text_list)
        r.used_flash_rodata = get_size(flash_rodata_list)

        r.used_flash_other = get_size(flash_other_list)
        r.used_flash = r.used_flash_text + r.used_flash_rodata + r.used_flash_other

        # The used DRAM BSS is counted into the "Used static DRAM" but not into the "Total image size"
        r.total_size = r.used_dram - r.used_dram_bss + r.used_iram + r.used_diram - r.used_diram_bss + r.used_flash
        return r

    def get_json_dic(self):  # type: (StructureForSummary) -> collections.OrderedDict
        ret = collections.OrderedDict([
            ('dram_data', self.used_dram_data),
            ('dram_bss', self.used_dram_bss),
            ('dram_rodata', self.used_dram_rodata),
            ('dram_other', self.used_dram_other),
            ('used_dram', self.used_dram),
            ('dram_total', self.dram_total),
            ('used_dram_ratio', self.used_dram_ratio if self.used_dram_ratio is not float('nan') else 0),
            ('dram_remain', self.dram_remain),

            ('iram_vectors', self.used_iram_vectors),
            ('iram_text', self.used_iram_text),
            ('iram_other', self.used_iram_other),
            ('used_iram', self.used_iram),
            ('iram_total', self.iram_total),
            ('used_iram_ratio', self.used_iram_ratio),
            ('iram_remain', self.iram_remain),

            ('diram_data', self.used_diram_data),
            ('diram_bss', self.used_diram_bss),
            ('diram_text', self.used_diram_text),
            ('diram_vectors', self.used_diram_vectors),
            ('diram_rodata', self.used_diram_rodata),
            ('diram_other', self.used_diram_other),
            ('diram_total', self.diram_total),
            ('used_diram', self.used_diram),
            ('used_diram_ratio', self.used_diram_ratio),
            ('diram_remain', self.diram_remain),

            ('flash_code', self.used_flash_text),
            ('flash_rodata', self.used_flash_rodata),
            ('flash_other', self.used_flash_other),
            ('used_flash_non_ram', self.used_flash),    # text/data in D/I RAM not included

            ('total_size', self.total_size)             # bss not included
        ])
        assert len(ret) == len(StructureForSummary.get_required_items())
        return ret


def get_summary(path, segments, sections, target,
                as_json=False,
                path_diff='', segments_diff=None, sections_diff=None, target_diff=''):
    # type: (str, Dict, Dict, str, bool, str, Optional[Dict], Optional[Dict], str) -> str
    if segments_diff is None:
        segments_diff = {}
    if sections_diff is None:
        sections_diff = {}

    mem_regions = MemRegions(target)
    segment_layout = mem_regions.fit_segments_into_regions(segments)
    section_layout = mem_regions.fit_sections_into_regions(LinkingSections.filter_sections(sections))
    current = StructureForSummary.get(segment_layout, section_layout)

    if path_diff:
        diff_en = True
        mem_regions_diff = MemRegions(target_diff)
        segment_layout_diff = mem_regions_diff.fit_segments_into_regions(segments_diff)
        section_layout_diff = mem_regions_diff.fit_sections_into_regions(LinkingSections.filter_sections(sections_diff))
        reference = StructureForSummary.get(segment_layout_diff, section_layout_diff)
    else:
        diff_en = False
        reference = StructureForSummary()

    if as_json:
        current_json_dic = current.get_json_dic()
        if diff_en:
            reference_json_dic = reference.get_json_dic()
            diff_json_dic = collections.OrderedDict([
                (k, v - reference_json_dic[k]) for k, v in iteritems(current_json_dic)])
            output = format_json(collections.OrderedDict([('current', current_json_dic),
                                                          ('reference', reference_json_dic),
                                                          ('diff', diff_json_dic),
                                                          ]))
        else:
            output = format_json(current_json_dic)
    else:
        class LineDef(object):
            title = ''
            name = ''

            def __init__(self, title, name):  # type: (LineDef, str, str) -> None
                self.title = title
                self.name = name

            def format_line(self):  # type: (LineDef) -> Tuple[str, str, str, str]
                return (self.title + ': {%s:>7} bytes' % self.name,
                        '{%s:>7}' % self.name,
                        '{%s:+}' % self.name,
                        '')

        class HeadLineDef(LineDef):
            remain = ''
            ratio = ''
            total = ''

            def __init__(self, title, name, remain, ratio, total):  # type: (HeadLineDef, str, str, str, str, str) -> None
                super(HeadLineDef, self).__init__(title, name)
                self.remain = remain
                self.ratio = ratio
                self.total = total

            def format_line(self):  # type: (HeadLineDef) -> Tuple[str, str, str, str]
                return ('%s: {%s:>7} bytes ({%s:>7} remain, {%s:.1%%} used)' % (self.title, self.name, self.remain, self.ratio),
                        '{%s:>7}' % self.name,
                        '{%s:+}' % self.name,
                        '({%s:>+7} remain, {%s:>+7} total)' % (self.remain, self.total))

        class TotalLineDef(LineDef):

            def format_line(self):  # type: (TotalLineDef) -> Tuple[str, str, str, str]
                return (self.title + ': {%s:>7} bytes (.bin may be padded larger)' % self.name,
                        '{%s:>7}' % self.name,
                        '{%s:+}' % self.name,
                        '')

        format_list = [
            HeadLineDef('Used static DRAM', 'used_dram', remain='dram_remain', ratio='used_dram_ratio', total='dram_total'),
            LineDef('      .data size', 'used_dram_data'),
            LineDef('      .bss  size', 'used_dram_bss'),
            LineDef('   .rodata  size', 'used_dram_rodata'),
            LineDef(' DRAM other size', 'used_dram_other'),

            HeadLineDef('Used static IRAM', 'used_iram', remain='iram_remain', ratio='used_iram_ratio', total='iram_total'),
            LineDef('      .text size', 'used_iram_text'),
            LineDef('   .vectors size', 'used_iram_vectors'),

            HeadLineDef('Used stat D/IRAM', 'used_diram', remain='diram_remain', ratio='used_diram_ratio', total='diram_total'),
            LineDef('      .data size', 'used_diram_data'),
            LineDef('      .bss  size', 'used_diram_bss'),
            LineDef('      .text size', 'used_diram_text'),
            LineDef('   .vectors size', 'used_diram_vectors'),
            LineDef('    .rodata size', 'used_diram_rodata'),
            LineDef('      other     ', 'used_diram_other'),

            LineDef('Used Flash size ', 'used_flash'),
            LineDef('      .text     ', 'used_flash_text'),
            LineDef('      .rodata   ', 'used_flash_rodata'),

            TotalLineDef('Total image size', 'total_size')
        ]

        def convert_to_fmt_dict(summary, suffix=''):  # type: (StructureForSummary, str) -> Dict
            required_items = StructureForSummary.get_required_items()
            return dict([(key + suffix, getattr(summary, key)) for key in required_items])

        f_dic1 = convert_to_fmt_dict(current)
        if diff_en:
            f_dic2 = convert_to_fmt_dict(reference)
            f_dic_diff = convert_to_fmt_dict(current - reference)

        lf = '{:60}{:>15}{:>15} {}'  # Width for a, b, c, d columns

        def print_in_columns(a, b='', c='', d=''):
            # type: (str, Optional[str], Optional[str], Optional[str]) -> str
            return lf.format(a, b, c, d).rstrip() + os.linesep

        output = ''
        if diff_en:
            output += print_in_columns('<CURRENT> MAP file: ' + path)
            output += print_in_columns('<REFERENCE> MAP file: ' + path_diff)
            output += print_in_columns('Difference is counted as <CURRENT> - <REFERENCE>, ',
                                       'i.e. a positive number means that <CURRENT> is larger.')
            output += print_in_columns('Total sizes of <CURRENT>:', '<REFERENCE>', 'Difference', '')

            for line in format_list:
                if getattr(current, line.name) > 0 or getattr(reference, line.name) > 0 or line.name == 'total_size':
                    a, b, c, d = line.format_line()
                    output += print_in_columns(
                        a.format(**f_dic1),
                        b.format(**f_dic2),
                        c.format(**f_dic_diff) if not c.format(**f_dic_diff).startswith('+0') else '',
                        d.format(**f_dic_diff))
        else:
            output += print_in_columns('Total sizes:')

            for line in format_list:
                if getattr(current, line.name) > 0 or line.name == 'total_size':
                    a, b, c, d = line.format_line()
                    output += print_in_columns(a.format(**f_dic1))

    return output


def check_is_dict_sort(non_sort_list):  # type: (List) -> List
    # keeping the order data, bss, other, iram, diram, ram_st_total, flash_text, flash_rodata, flash_total
    start_of_other = 0
    props_sort = []  # type: List
    props_elem = ['.data', '.bss', 'other', 'iram', 'diram', 'ram_st_total', 'flash.text', 'flash.rodata', 'flash', 'flash_total']
    for i in props_elem:
        for j in non_sort_list:
            if i == 'other':
                start_of_other = len(props_sort)
            elif i in (j[0] if len(j[0]) > 1 else j) and (j[0] if len(j[0]) > 1 else j) not in props_sort:
                props_sort.append(j)
    for j in non_sort_list:
        if j not in props_sort:
            props_sort.insert(start_of_other, j)
    return props_sort


class StructureForDetailedSizes(object):

    @staticmethod
    def sizes_by_key(sections, key, include_padding=False):  # type: (SectionDict, str, Optional[bool]) -> Dict[str, Dict[str, int]]
        """ Takes a dict of sections (from load_sections) and returns
        a dict keyed by 'key' with aggregate output size information.

        Key can be either "archive" (for per-archive data) or "file" (for per-file data) in the result.
        """
        result = {}  # type: Dict[str, Dict[str, int]]
        for _, section in iteritems(sections):
            for s in section['sources']:
                if not s[key] in result:
                    result[s[key]] = {}
                archive = result[s[key]]
                if not section['name'] in archive:
                    archive[section['name']] = 0
                archive[section['name']] += s['size']
                if include_padding:
                    archive[section['name']] += s['fill']
        return result

    @staticmethod
    def get(sections, by_key):  # type: (SectionDict, str) -> collections.OrderedDict
        # Get the detailed structure before using the filter to remove undesired sections,
        # to show entries without desired sections
        sizes = StructureForDetailedSizes.sizes_by_key(sections, by_key)
        for key_name in sizes:
            sizes[key_name] = LinkingSections.filter_sections(sizes[key_name])

        s = []
        for key, section_dict in sizes.items():
            ram_st_total = sum([x[1] for x in section_dict.items() if not LinkingSections.in_section(x[0], 'flash')])
            flash_total = sum([x[1] for x in section_dict.items() if not LinkingSections.in_section(x[0], 'bss')])  # type: int

            section_dict['ram_st_total'] = ram_st_total
            section_dict['flash_total'] = flash_total

            sorted_dict = sorted(section_dict.items(), key=lambda elem: elem[0])
            sorted_dict = check_is_dict_sort(sorted_dict)

            s.append((key, collections.OrderedDict(sorted_dict)))

        s = sorted(s, key=lambda elem: elem[0])
        # do a secondary sort in order to have consistent order (for diff-ing the output)
        s = sorted(s, key=lambda elem: elem[1]['flash_total'], reverse=True)

        return collections.OrderedDict(s)


def get_detailed_sizes(sections, key, header, as_json=False, sections_diff=None):  # type: (Dict, str, str, bool, Dict) -> str

    key_name_set = set()
    current = StructureForDetailedSizes.get(sections, key)
    for section_dict in current.values():
        key_name_set.update(section_dict.keys())

    if sections_diff:
        reference = StructureForDetailedSizes.get(sections_diff, key)
        for section_dict in reference.values():
            key_name_set.update(section_dict.keys())
        diff_en = True
    else:
        diff_en = False

    key_name_list = list(key_name_set)
    ordered_key_list, display_name_list = LinkingSections.get_display_name_order(key_name_list)

    if as_json:
        if diff_en:
            diff_json_dic = collections.OrderedDict()
            for name in sorted(list(frozenset(current.keys()) | frozenset(reference.keys()))):
                cur_name_dic = current.get(name, {})
                ref_name_dic = reference.get(name, {})
                all_keys = sorted(list(frozenset(cur_name_dic.keys()) | frozenset(ref_name_dic.keys())))
                diff_json_dic[name] = collections.OrderedDict([(k,
                                                                cur_name_dic.get(k, 0) -
                                                                ref_name_dic.get(k, 0)) for k in all_keys])
            output = format_json(collections.OrderedDict([('current', current),
                                                          ('reference', reference),
                                                          ('diff', diff_json_dic),
                                                          ]))
        else:
            output = format_json(current)
    else:
        def _get_header_format(disp_list=display_name_list):  # type: (List) -> str
            len_list = [len(x) for x in disp_list]
            len_list.insert(0, 24)
            return ' '.join(['{:>%d}' % x for x in len_list]) + os.linesep

        def _get_output(data, selection, key_list=ordered_key_list, disp_list=display_name_list):
            # type: (Dict[str, Dict[str, int]], Collection, List, List) -> str
            header_format = _get_header_format(disp_list)
            output = header_format.format(header, *disp_list)

            for k, v in iteritems(data):
                if k not in selection:
                    continue

                try:
                    _, k = k.split(':', 1)
                    # print subheadings for key of format archive:file
                except ValueError:
                    # k remains the same
                    pass

                def get_section_size(section_dict):  # type: (Dict) -> Callable[[str], int]
                    return lambda x: section_dict.get(x, 0)

                section_size_list = map(get_section_size(section_dict=v), key_list)
                output += header_format.format(k[:24], *(section_size_list))
            return output

        def _get_header_format_diff(disp_list=display_name_list, columns=False):  # type: (List, bool) -> str
            if columns:
                len_list = (24, ) + (7, ) * 3 * len(disp_list)
                return '|'.join(['{:>%d}' % x for x in len_list]) + os.linesep

            len_list = (24, ) + (23, ) * len(disp_list)
            return ' '.join(['{:>%d}' % x for x in len_list]) + os.linesep

        def _get_output_diff(curr, ref, key_list=ordered_key_list, disp_list=display_name_list):
            # type: (Dict, Dict, List, List) -> str
            # First header without Current/Ref/Diff columns
            header_format = _get_header_format_diff(columns=False)
            output = header_format.format(header, *disp_list)

            f_print = ('-' * 23, '') * len(key_list)
            f_print = f_print[0:len(key_list)]
            header_line = header_format.format('', *f_print)

            header_format = _get_header_format_diff(columns=True)
            f_print = ('<C>', '<R>', '<C>-<R>') * len(key_list)

            output += header_format.format('', *f_print)
            output += header_line

            for k, v in iteritems(curr):
                try:
                    v2 = ref[k]
                except KeyError:
                    continue

                try:
                    _, k = k.split(':', 1)
                    # print subheadings for key of format archive:file
                except ValueError:
                    # k remains the same
                    pass

                def _get_items(name, section_dict=v, section_dict_ref=v2):
                    # type: (str, Dict, Dict) -> Tuple[str, str, str]
                    a = section_dict.get(name, 0)
                    b = section_dict_ref.get(name, 0)
                    diff = a - b
                    # the sign is added here and not in header_format in order to be able to print empty strings
                    return (a or '', b or '', '' if diff == 0 else '{:+}'.format(diff))

                x = []  # type: List[str]
                for section in key_list:
                    x.extend(_get_items(section))

                output += header_format.format(k[:24], *(x))

            return output

        output = 'Per-{} contributions to ELF file:{}'.format(key, os.linesep)

        if diff_en:
            output += _get_output_diff(current, reference)

            in_current = frozenset(current.keys())
            in_reference = frozenset(reference.keys())
            only_in_current = in_current - in_reference
            only_in_reference = in_reference - in_current

            if len(only_in_current) > 0:
                output += 'The following entries are present in <CURRENT> only:{}'.format(os.linesep)
                output += _get_output(current, only_in_current)

            if len(only_in_reference) > 0:
                output += 'The following entries are present in <REFERENCE> only:{}'.format(os.linesep)
                output += _get_output(reference, only_in_reference)
        else:
            output += _get_output(current, current)

    return output


class StructureForArchiveSymbols(object):
    @staticmethod
    def get(archive, sections):  # type: (str, Dict) -> Dict
        interested_sections = LinkingSections.filter_sections(sections)

        result = dict([(t, {}) for t in interested_sections])  # type: Dict[str, Dict[str, int]]
        for _, section in iteritems(sections):
            section_name = section['name']
            if section_name not in interested_sections:
                continue
            for s in section['sources']:
                if archive != s['archive']:
                    continue
                s['sym_name'] = re.sub('(.text.|.literal.|.data.|.bss.|.rodata.)', '', s['sym_name'])
                result[section_name][s['sym_name']] = result[section_name].get(s['sym_name'], 0) + s['size']

        # build a new ordered dict of each section, where each entry is an ordereddict of symbols to sizes
        section_symbols = collections.OrderedDict()
        for t in sorted(list(interested_sections)):
            s = sorted(result[t].items(), key=lambda k_v: str(k_v[0]))
            # do a secondary sort in order to have consistent order (for diff-ing the output)
            s = sorted(s, key=lambda k_v: int(k_v[1]), reverse=True)
            section_symbols[t] = collections.OrderedDict(s)

        return section_symbols


def get_archive_symbols(sections, archive, as_json=False, sections_diff=None):  # type: (Dict, str, bool, Dict) -> str
    diff_en = bool(sections_diff)
    current = StructureForArchiveSymbols.get(archive, sections)
    reference = StructureForArchiveSymbols.get(archive, sections_diff) if sections_diff else {}

    if as_json:
        if diff_en:
            diff_json_dic = collections.OrderedDict()
            for name in sorted(list(frozenset(current.keys()) | frozenset(reference.keys()))):
                cur_name_dic = current.get(name, {})
                ref_name_dic = reference.get(name, {})
                all_keys = sorted(list(frozenset(cur_name_dic.keys()) | frozenset(ref_name_dic.keys())))
                diff_json_dic[name] = collections.OrderedDict([(key,
                                                                cur_name_dic.get(key, 0) -
                                                                ref_name_dic.get(key, 0)) for key in all_keys])
            output = format_json(collections.OrderedDict([('current', current),
                                                          ('reference', reference),
                                                          ('diff', diff_json_dic),
                                                          ]))
        else:
            output = format_json(current)
    else:
        def _get_item_pairs(name, section):  # type: (str, collections.OrderedDict) -> collections.OrderedDict
            return collections.OrderedDict([(key.replace(name + '.', ''), val) for key, val in iteritems(section)])

        def _get_max_len(symbols_dict):  # type: (Dict) -> Tuple[int, int]
            names_max_len = 0
            numbers_max_len = 0
            for t, s in iteritems(symbols_dict):
                numbers_max_len = max([numbers_max_len] + [len(str(x)) for _, x in iteritems(s)])
                names_max_len = max([names_max_len] + [len(x) for x in _get_item_pairs(t, s)])

            return names_max_len, numbers_max_len

        def _get_output(section_symbols):  # type: (Dict) -> str
            output = ''
            names_max_len, numbers_max_len  = _get_max_len(section_symbols)
            for t, s in iteritems(section_symbols):
                output += '{}Symbols from section: {}{}'.format(os.linesep, t, os.linesep)
                item_pairs = _get_item_pairs(t, s)
                for key, val in iteritems(item_pairs):
                    output += ' '.join([('\t{:<%d} : {:>%d}\n' % (names_max_len,numbers_max_len)).format(key, val)])
                section_total = sum([val for _, val in iteritems(item_pairs)])
                output += 'Section total: {}{}'.format(section_total, os.linesep)
            return output

        output = '{}Symbols within the archive: {} (Not all symbols may be reported){}'.format(os.linesep, archive, os.linesep)
        if diff_en:

            def _generate_line_tuple(curr, ref, name):
                # type: (collections.OrderedDict, collections.OrderedDict, str) -> Tuple[str, int, int, str]
                cur_val = curr.get(name, 0)
                ref_val = ref.get(name, 0)
                diff_val = cur_val - ref_val
                # string slicing is used just to make sure it will fit into the first column of line_format
                return ((' ' * 4 + name)[:40], cur_val, ref_val, '' if diff_val == 0 else '{:+}'.format(diff_val))

            line_format = '{:40} {:>12} {:>12} {:>25}'
            all_section_names = sorted(list(frozenset(current.keys()) | frozenset(reference.keys())))
            for section_name in all_section_names:
                current_item_pairs = _get_item_pairs(section_name, current.get(section_name, {}))
                reference_item_pairs = _get_item_pairs(section_name, reference.get(section_name, {}))
                output += os.linesep + line_format.format(section_name[:40],
                                                          '<CURRENT>',
                                                          '<REFERENCE>',
                                                          '<CURRENT> - <REFERENCE>') + os.linesep
                current_section_total = sum([val for _, val in iteritems(current_item_pairs)])
                reference_section_total = sum([val for _, val in iteritems(reference_item_pairs)])
                diff_section_total = current_section_total - reference_section_total
                all_item_names = sorted(list(frozenset(current_item_pairs.keys()) |
                                             frozenset(reference_item_pairs.keys())))
                output += os.linesep.join([line_format.format(*_generate_line_tuple(current_item_pairs,
                                                                                    reference_item_pairs,
                                                                                    n)
                                                              ).rstrip() for n in all_item_names])
                output += os.linesep if current_section_total > 0 or reference_section_total > 0 else ''
                output += line_format.format('Section total:',
                                             current_section_total,
                                             reference_section_total,
                                             '' if diff_section_total == 0 else '{:+}'.format(diff_section_total)
                                             ).rstrip() + os.linesep
        else:
            output += _get_output(current)
    return output


if __name__ == '__main__':
    main()