xref: /aosp_15_r20/system/extras/simpleperf/scripts/report_html.py (revision 288bf5226967eb3dac5cce6c939ccc2a7f2b4fe5)

#!/usr/bin/env python3
#
# Copyright (C) 2017 The Android Open Source Project
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#

from __future__ import annotations
import argparse
import collections
from concurrent.futures import Future, ThreadPoolExecutor
from dataclasses import dataclass
import datetime
import json
import logging
import os
from pathlib import Path
import sys
from typing import Any, Callable, Dict, Iterator, List, Optional, Set, Tuple, Union

from simpleperf_report_lib import GetReportLib, SymbolStruct
from simpleperf_utils import (
    Addr2Nearestline, AddrRange, BaseArgumentParser, BinaryFinder, Disassembly, get_script_dir,
    log_exit, Objdump, open_report_in_browser, ReadElf, ReportLibOptions, SourceFileSearcher)

MAX_CALLSTACK_LENGTH = 750


class HtmlWriter(object):

    def __init__(self, output_path: Union[Path, str]):
        self.fh = open(output_path, 'w')
        self.tag_stack = []

    def close(self):
        self.fh.close()

    def open_tag(self, tag: str, **attrs: Dict[str, str]) -> HtmlWriter:
        attr_str = ''
        for key in attrs:
            attr_str += ' %s="%s"' % (key, attrs[key])
        self.fh.write('<%s%s>' % (tag, attr_str))
        self.tag_stack.append(tag)
        return self

    def close_tag(self, tag: Optional[str] = None):
        if tag:
            assert tag == self.tag_stack[-1]
        self.fh.write('</%s>\n' % self.tag_stack.pop())

    def add(self, text: str) -> HtmlWriter:
        self.fh.write(text)
        return self

    def add_file(self, file_path: Union[Path, str]) -> HtmlWriter:
        file_path = os.path.join(get_script_dir(), file_path)
        with open(file_path, 'r') as f:
            self.add(f.read())
        return self


def modify_text_for_html(text: str) -> str:
    return text.replace('>', '&gt;').replace('<', '&lt;')


def hex_address_for_json(addr: int) -> str:
    """ To handle big addrs (nears uint64_max) in Javascript, store addrs as hex strings in Json.
    """
    return '0x%x' % addr


class EventScope(object):

    def __init__(self, name: str):
        self.name = name
        self.processes: Dict[int, ProcessScope] = {}  # map from pid to ProcessScope
        self.sample_count = 0
        self.event_count = 0

    def get_process(self, pid: int) -> ProcessScope:
        process = self.processes.get(pid)
        if not process:
            process = self.processes[pid] = ProcessScope(pid)
        return process

    def get_sample_info(self, gen_addr_hit_map: bool) -> Dict[str, Any]:
        result = {}
        result['eventName'] = self.name
        result['eventCount'] = self.event_count
        processes = sorted(self.processes.values(), key=lambda a: a.event_count, reverse=True)
        result['processes'] = [process.get_sample_info(gen_addr_hit_map)
                               for process in processes]
        return result

    @property
    def threads(self) -> Iterator[ThreadScope]:
        for process in self.processes.values():
            for thread in process.threads.values():
                yield thread

    @property
    def libraries(self) -> Iterator[LibScope]:
        for process in self.processes.values():
            for thread in process.threads.values():
                for lib in thread.libs.values():
                    yield lib


class ProcessScope(object):

    def __init__(self, pid: int):
        self.pid = pid
        self.name = ''
        self.event_count = 0
        self.threads: Dict[int, ThreadScope] = {}  # map from tid to ThreadScope

    def get_thread(self, tid: int, thread_name: str) -> ThreadScope:
        thread = self.threads.get(tid)
        if not thread:
            thread = self.threads[tid] = ThreadScope(tid)
        thread.name = thread_name
        if self.pid == tid:
            self.name = thread_name
        return thread

    def get_sample_info(self, gen_addr_hit_map: bool) -> Dict[str, Any]:
        result = {}
        result['pid'] = self.pid
        result['eventCount'] = self.event_count
        # Sorting threads by sample count is better for profiles recorded with --trace-offcpu.
        threads = sorted(self.threads.values(), key=lambda a: a.sample_count, reverse=True)
        result['threads'] = [thread.get_sample_info(gen_addr_hit_map)
                             for thread in threads]
        return result

    def merge_by_thread_name(self, process: ProcessScope):
        self.event_count += process.event_count
        thread_list: List[ThreadScope] = list(
            self.threads.values()) + list(process.threads.values())
        new_threads: Dict[str, ThreadScope] = {}  # map from thread name to ThreadScope
        for thread in thread_list:
            cur_thread = new_threads.get(thread.name)
            if cur_thread is None:
                new_threads[thread.name] = thread
            else:
                cur_thread.merge(thread)
        self.threads = {}
        for thread in new_threads.values():
            self.threads[thread.tid] = thread


class ThreadScope(object):

    def __init__(self, tid: int):
        self.tid = tid
        self.name = ''
        self.event_count = 0
        self.sample_count = 0
        self.libs: Dict[int, LibScope] = {}  # map from lib_id to LibScope
        self.call_graph = CallNode(-1)
        self.reverse_call_graph = CallNode(-1)

    def add_callstack(
            self, event_count: int, callstack: List[Tuple[int, int, int]],
            build_addr_hit_map: bool):
        """ callstack is a list of tuple (lib_id, func_id, addr).
            For each i > 0, callstack[i] calls callstack[i-1]."""
        hit_func_ids: Set[int] = set()
        for i, (lib_id, func_id, addr) in enumerate(callstack):
            # When a callstack contains recursive function, only add for each function once.
            if func_id in hit_func_ids:
                continue
            hit_func_ids.add(func_id)

            lib = self.libs.get(lib_id)
            if not lib:
                lib = self.libs[lib_id] = LibScope(lib_id)
            function = lib.get_function(func_id)
            function.subtree_event_count += event_count
            if i == 0:
                lib.event_count += event_count
                function.event_count += event_count
                function.sample_count += 1
            if build_addr_hit_map:
                function.build_addr_hit_map(addr, event_count if i == 0 else 0, event_count)

        # build call graph and reverse call graph
        node = self.call_graph
        for item in reversed(callstack):
            node = node.get_child(item[1])
        node.event_count += event_count
        node = self.reverse_call_graph
        for item in callstack:
            node = node.get_child(item[1])
        node.event_count += event_count

    def update_subtree_event_count(self):
        self.call_graph.update_subtree_event_count()
        self.reverse_call_graph.update_subtree_event_count()

    def limit_percents(self, min_func_limit: float, min_callchain_percent: float,
                       hit_func_ids: Set[int]):
        for lib in self.libs.values():
            to_del_funcs = []
            for function in lib.functions.values():
                if function.subtree_event_count < min_func_limit:
                    to_del_funcs.append(function.func_id)
                else:
                    hit_func_ids.add(function.func_id)
            for func_id in to_del_funcs:
                del lib.functions[func_id]
        min_limit = min_callchain_percent * 0.01 * self.call_graph.subtree_event_count
        self.call_graph.cut_edge(min_limit, hit_func_ids)
        self.reverse_call_graph.cut_edge(min_limit, hit_func_ids)

    def get_sample_info(self, gen_addr_hit_map: bool) -> Dict[str, Any]:
        result = {}
        result['tid'] = self.tid
        result['eventCount'] = self.event_count
        result['sampleCount'] = self.sample_count
        result['libs'] = [lib.gen_sample_info(gen_addr_hit_map)
                          for lib in self.libs.values()]
        result['g'] = self.call_graph.gen_sample_info()
        result['rg'] = self.reverse_call_graph.gen_sample_info()
        return result

    def merge(self, thread: ThreadScope):
        self.event_count += thread.event_count
        self.sample_count += thread.sample_count
        for lib_id, lib in thread.libs.items():
            cur_lib = self.libs.get(lib_id)
            if cur_lib is None:
                self.libs[lib_id] = lib
            else:
                cur_lib.merge(lib)
        self.call_graph.merge(thread.call_graph)
        self.reverse_call_graph.merge(thread.reverse_call_graph)

    def sort_call_graph_by_function_name(self, get_func_name: Callable[[int], str]) -> None:
        self.call_graph.sort_by_function_name(get_func_name)
        self.reverse_call_graph.sort_by_function_name(get_func_name)


class LibScope(object):

    def __init__(self, lib_id: int):
        self.lib_id = lib_id
        self.event_count = 0
        self.functions: Dict[int, FunctionScope] = {}  # map from func_id to FunctionScope.

    def get_function(self, func_id: int) -> FunctionScope:
        function = self.functions.get(func_id)
        if not function:
            function = self.functions[func_id] = FunctionScope(func_id)
        return function

    def gen_sample_info(self, gen_addr_hit_map: bool) -> Dict[str, Any]:
        result = {}
        result['libId'] = self.lib_id
        result['eventCount'] = self.event_count
        result['functions'] = [func.gen_sample_info(gen_addr_hit_map)
                               for func in self.functions.values()]
        return result

    def merge(self, lib: LibScope):
        self.event_count += lib.event_count
        for func_id, function in lib.functions.items():
            cur_function = self.functions.get(func_id)
            if cur_function is None:
                self.functions[func_id] = function
            else:
                cur_function.merge(function)


class FunctionScope(object):

    def __init__(self, func_id: int):
        self.func_id = func_id
        self.sample_count = 0
        self.event_count = 0
        self.subtree_event_count = 0
        self.addr_hit_map = None  # map from addr to [event_count, subtree_event_count].
        # map from (source_file_id, line) to [event_count, subtree_event_count].
        self.line_hit_map = None

    def build_addr_hit_map(self, addr: int, event_count: int, subtree_event_count: int):
        if self.addr_hit_map is None:
            self.addr_hit_map = {}
        count_info = self.addr_hit_map.get(addr)
        if count_info is None:
            self.addr_hit_map[addr] = [event_count, subtree_event_count]
        else:
            count_info[0] += event_count
            count_info[1] += subtree_event_count

    def build_line_hit_map(self, source_file_id: int, line: int, event_count: int,
                           subtree_event_count: int):
        if self.line_hit_map is None:
            self.line_hit_map = {}
        key = (source_file_id, line)
        count_info = self.line_hit_map.get(key)
        if count_info is None:
            self.line_hit_map[key] = [event_count, subtree_event_count]
        else:
            count_info[0] += event_count
            count_info[1] += subtree_event_count

    def gen_sample_info(self, gen_addr_hit_map: bool) -> Dict[str, Any]:
        result = {}
        result['f'] = self.func_id
        result['c'] = [self.sample_count, self.event_count, self.subtree_event_count]
        if self.line_hit_map:
            items = []
            for key in self.line_hit_map:
                count_info = self.line_hit_map[key]
                item = {'f': key[0], 'l': key[1], 'e': count_info[0], 's': count_info[1]}
                items.append(item)
            result['s'] = items
        if gen_addr_hit_map and self.addr_hit_map:
            items = []
            for addr in sorted(self.addr_hit_map):
                count_info = self.addr_hit_map[addr]
                items.append(
                    {'a': hex_address_for_json(addr),
                     'e': count_info[0],
                     's': count_info[1]})
            result['a'] = items
        return result

    def merge(self, function: FunctionScope):
        self.sample_count += function.sample_count
        self.event_count += function.event_count
        self.subtree_event_count += function.subtree_event_count
        self.addr_hit_map = self.__merge_hit_map(self.addr_hit_map, function.addr_hit_map)
        self.line_hit_map = self.__merge_hit_map(self.line_hit_map, function.line_hit_map)

    @staticmethod
    def __merge_hit_map(map1: Optional[Dict[int, List[int]]],
                        map2: Optional[Dict[int, List[int]]]) -> Optional[Dict[int, List[int]]]:
        if not map1:
            return map2
        if not map2:
            return map1
        for key, value2 in map2.items():
            value1 = map1.get(key)
            if value1 is None:
                map1[key] = value2
            else:
                value1[0] += value2[0]
                value1[1] += value2[1]
        return map1


class CallNode(object):

    def __init__(self, func_id: int):
        self.event_count = 0
        self.subtree_event_count = 0
        self.func_id = func_id
        # map from func_id to CallNode
        self.children: Dict[int, CallNode] = collections.OrderedDict()

    def get_child(self, func_id: int) -> CallNode:
        child = self.children.get(func_id)
        if not child:
            child = self.children[func_id] = CallNode(func_id)
        return child

    def update_subtree_event_count(self):
        self.subtree_event_count = self.event_count
        for child in self.children.values():
            self.subtree_event_count += child.update_subtree_event_count()
        return self.subtree_event_count

    def cut_edge(self, min_limit: float, hit_func_ids: Set[int]):
        hit_func_ids.add(self.func_id)
        to_del_children = []
        for key in self.children:
            child = self.children[key]
            if child.subtree_event_count < min_limit:
                to_del_children.append(key)
            else:
                child.cut_edge(min_limit, hit_func_ids)
        for key in to_del_children:
            del self.children[key]

    def gen_sample_info(self) -> Dict[str, Any]:
        result = {}
        result['e'] = self.event_count
        result['s'] = self.subtree_event_count
        result['f'] = self.func_id
        result['c'] = [child.gen_sample_info() for child in self.children.values()]
        return result

    def merge(self, node: CallNode):
        self.event_count += node.event_count
        self.subtree_event_count += node.subtree_event_count
        for key, child in node.children.items():
            cur_child = self.children.get(key)
            if cur_child is None:
                self.children[key] = child
            else:
                cur_child.merge(child)

    def sort_by_function_name(self, get_func_name: Callable[[int], str]) -> None:
        if self.children:
            child_func_ids = list(self.children.keys())
            child_func_ids.sort(key=get_func_name)
            new_children = collections.OrderedDict()
            for func_id in child_func_ids:
                new_children[func_id] = self.children[func_id]
            self.children = new_children
            for child in self.children.values():
                child.sort_by_function_name(get_func_name)


@dataclass
class LibInfo:
    name: str
    build_id: str


class LibSet(object):
    """ Collection of shared libraries used in perf.data. """

    def __init__(self):
        self.lib_name_to_id: Dict[str, int] = {}
        self.libs: List[LibInfo] = []

    def get_lib_id(self, lib_name: str) -> Optional[int]:
        return self.lib_name_to_id.get(lib_name)

    def add_lib(self, lib_name: str, build_id: str) -> int:
        """ Return lib_id of the newly added lib. """
        lib_id = len(self.libs)
        self.libs.append(LibInfo(lib_name, build_id))
        self.lib_name_to_id[lib_name] = lib_id
        return lib_id

    def get_lib(self, lib_id: int) -> LibInfo:
        return self.libs[lib_id]


class Function(object):
    """ Represent a function in a shared library. """

    def __init__(self, lib_id: int, func_name: str, func_id: int, start_addr: int, addr_len: int):
        self.lib_id = lib_id
        self.func_name = func_name
        self.func_id = func_id
        self.start_addr = start_addr
        self.addr_len = addr_len
        self.source_info = None
        self.disassembly = None


class FunctionSet(object):
    """ Collection of functions used in perf.data. """

    def __init__(self):
        self.name_to_func: Dict[Tuple[int, str], Function] = {}
        self.id_to_func: Dict[int, Function] = {}

    def get_func_id(self, lib_id: int, symbol: SymbolStruct) -> int:
        key = (lib_id, symbol.symbol_name)
        function = self.name_to_func.get(key)
        if function is None:
            func_id = len(self.id_to_func)
            function = Function(lib_id, symbol.symbol_name, func_id, symbol.symbol_addr,
                                symbol.symbol_len)
            self.name_to_func[key] = function
            self.id_to_func[func_id] = function
        return function.func_id

    def get_func_name(self, func_id: int) -> str:
        return self.id_to_func[func_id].func_name

    def trim_functions(self, left_func_ids: Set[int]):
        """ Remove functions excepts those in left_func_ids. """
        for function in self.name_to_func.values():
            if function.func_id not in left_func_ids:
                del self.id_to_func[function.func_id]
        # name_to_func will not be used.
        self.name_to_func = None


class SourceFile(object):
    """ A source file containing source code hit by samples. """

    def __init__(self, file_id: int, abstract_path: str):
        self.file_id = file_id
        self.abstract_path = abstract_path  # path reported by addr2line
        self.real_path: Optional[str] = None  # file path in the file system
        self.requested_lines: Optional[Set[int]] = set()
        self.line_to_code: Dict[int, str] = {}  # map from line to code in that line.

    def request_lines(self, start_line: int, end_line: int):
        self.requested_lines |= set(range(start_line, end_line + 1))

    def add_source_code(self, real_path: str):
        self.real_path = real_path
        with open(real_path, 'r') as f:
            source_code = f.readlines()
        max_line = len(source_code)
        for line in self.requested_lines:
            if line > 0 and line <= max_line:
                self.line_to_code[line] = source_code[line - 1]
        # requested_lines is no longer used.
        self.requested_lines = None


class SourceFileSet(object):
    """ Collection of source files. """

    def __init__(self):
        self.path_to_source_files: Dict[str, SourceFile] = {}  # map from file path to SourceFile.

    def get_source_file(self, file_path: str) -> SourceFile:
        source_file = self.path_to_source_files.get(file_path)
        if not source_file:
            source_file = SourceFile(len(self.path_to_source_files), file_path)
            self.path_to_source_files[file_path] = source_file
        return source_file

    def load_source_code(self, source_dirs: List[str]):
        file_searcher = SourceFileSearcher(source_dirs)
        for source_file in self.path_to_source_files.values():
            real_path = file_searcher.get_real_path(source_file.abstract_path)
            if real_path:
                source_file.add_source_code(real_path)


class RecordData(object):

    """RecordData reads perf.data, and generates data used by report_html.js in json format.
        All generated items are listed as below:
            1. recordTime: string
            2. machineType: string
            3. androidVersion: string
            4. recordCmdline: string
            5. totalSamples: int
            6. processNames: map from pid to processName.
            7. threadNames: map from tid to threadName.
            8. libList: an array of libNames, indexed by libId.
            9. functionMap: map from functionId to funcData.
                funcData = {
                    l: libId
                    f: functionName
                    s: [sourceFileId, startLine, endLine] [optional]
                    d: [(disassembly, addr)] [optional]
                }

            10.  sampleInfo = [eventInfo]
                eventInfo = {
                    eventName
                    eventCount
                    processes: [processInfo]
                }
                processInfo = {
                    pid
                    eventCount
                    threads: [threadInfo]
                }
                threadInfo = {
                    tid
                    eventCount
                    sampleCount
                    libs: [libInfo],
                    g: callGraph,
                    rg: reverseCallgraph
                }
                libInfo = {
                    libId,
                    eventCount,
                    functions: [funcInfo]
                }
                funcInfo = {
                    f: functionId
                    c: [sampleCount, eventCount, subTreeEventCount]
                    s: [sourceCodeInfo] [optional]
                    a: [addrInfo] (sorted by addrInfo.addr) [optional]
                }
                callGraph and reverseCallGraph are both of type CallNode.
                callGraph shows how a function calls other functions.
                reverseCallGraph shows how a function is called by other functions.
                CallNode {
                    e: selfEventCount
                    s: subTreeEventCount
                    f: functionId
                    c: [CallNode] # children
                }

                sourceCodeInfo {
                    f: sourceFileId
                    l: line
                    e: eventCount
                    s: subtreeEventCount
                }

                addrInfo {
                    a: addr
                    e: eventCount
                    s: subtreeEventCount
                }

            11. sourceFiles: an array of sourceFile, indexed by sourceFileId.
                sourceFile {
                    path
                    code:  # a map from line to code for that line.
                }
    """

    def __init__(
            self, binary_cache_path: Optional[str],
            ndk_path: Optional[str],
            build_addr_hit_map: bool):
        self.binary_cache_path = binary_cache_path
        self.ndk_path = ndk_path
        self.build_addr_hit_map = build_addr_hit_map
        self.meta_info: Optional[Dict[str, str]] = None
        self.cmdline: Optional[str] = None
        self.arch: Optional[str] = None
        self.events: Dict[str, EventScope] = {}
        self.libs = LibSet()
        self.functions = FunctionSet()
        self.total_samples = 0
        self.source_files = SourceFileSet()
        self.gen_addr_hit_map_in_record_info = False
        self.binary_finder = BinaryFinder(binary_cache_path, ReadElf(ndk_path))

    def load_record_file(self, record_file: str, report_lib_options: ReportLibOptions):
        lib = GetReportLib(record_file)
        # If not showing ip for unknown symbols, the percent of the unknown symbol may be
        # accumulated to very big, and ranks first in the sample table.
        lib.ShowIpForUnknownSymbol()
        if self.binary_cache_path:
            lib.SetSymfs(self.binary_cache_path)
        lib.SetReportOptions(report_lib_options)
        self.meta_info = lib.MetaInfo()
        self.cmdline = lib.GetRecordCmd()
        self.arch = lib.GetArch()
        while True:
            raw_sample = lib.GetNextSample()
            if not raw_sample:
                lib.Close()
                break
            raw_event = lib.GetEventOfCurrentSample()
            symbol = lib.GetSymbolOfCurrentSample()
            callchain = lib.GetCallChainOfCurrentSample()
            event = self._get_event(raw_event.name)
            self.total_samples += 1
            event.sample_count += 1
            event.event_count += raw_sample.period
            process = event.get_process(raw_sample.pid)
            process.event_count += raw_sample.period
            thread = process.get_thread(raw_sample.tid, raw_sample.thread_comm)
            thread.event_count += raw_sample.period
            thread.sample_count += 1

            lib_id = self.libs.get_lib_id(symbol.dso_name)
            if lib_id is None:
                lib_id = self.libs.add_lib(symbol.dso_name, lib.GetBuildIdForPath(symbol.dso_name))
            func_id = self.functions.get_func_id(lib_id, symbol)
            callstack = [(lib_id, func_id, symbol.vaddr_in_file)]
            for i in range(callchain.nr):
                symbol = callchain.entries[i].symbol
                lib_id = self.libs.get_lib_id(symbol.dso_name)
                if lib_id is None:
                    lib_id = self.libs.add_lib(
                        symbol.dso_name, lib.GetBuildIdForPath(symbol.dso_name))
                func_id = self.functions.get_func_id(lib_id, symbol)
                callstack.append((lib_id, func_id, symbol.vaddr_in_file))
            if len(callstack) > MAX_CALLSTACK_LENGTH:
                callstack = callstack[:MAX_CALLSTACK_LENGTH]
            thread.add_callstack(raw_sample.period, callstack, self.build_addr_hit_map)

        for event in self.events.values():
            for thread in event.threads:
                thread.update_subtree_event_count()

    def aggregate_by_thread_name(self):
        for event in self.events.values():
            new_processes = {}  # from process name to ProcessScope
            for process in event.processes.values():
                cur_process = new_processes.get(process.name)
                if cur_process is None:
                    new_processes[process.name] = process
                else:
                    cur_process.merge_by_thread_name(process)
            event.processes = {}
            for process in new_processes.values():
                event.processes[process.pid] = process

    def limit_percents(self, min_func_percent: float, min_callchain_percent: float):
        hit_func_ids: Set[int] = set()
        for event in self.events.values():
            min_limit = event.event_count * min_func_percent * 0.01
            to_del_processes = []
            for process in event.processes.values():
                to_del_threads = []
                for thread in process.threads.values():
                    if thread.call_graph.subtree_event_count < min_limit:
                        to_del_threads.append(thread.tid)
                    else:
                        thread.limit_percents(min_limit, min_callchain_percent, hit_func_ids)
                for thread in to_del_threads:
                    del process.threads[thread]
                if not process.threads:
                    to_del_processes.append(process.pid)
            for process in to_del_processes:
                del event.processes[process]
        self.functions.trim_functions(hit_func_ids)

    def sort_call_graph_by_function_name(self) -> None:
        for event in self.events.values():
            for process in event.processes.values():
                for thread in process.threads.values():
                    thread.sort_call_graph_by_function_name(self.functions.get_func_name)

    def _get_event(self, event_name: str) -> EventScope:
        if event_name not in self.events:
            self.events[event_name] = EventScope(event_name)
        return self.events[event_name]

    def add_source_code(self, source_dirs: List[str], filter_lib: Callable[[str], bool], jobs: int):
        """ Collect source code information:
            1. Find line ranges for each function in FunctionSet.
            2. Find line for each addr in FunctionScope.addr_hit_map.
            3. Collect needed source code in SourceFileSet.
        """
        addr2line = Addr2Nearestline(self.ndk_path, self.binary_finder, False)
        # Request line range for each function.
        for function in self.functions.id_to_func.values():
            if function.func_name == 'unknown':
                continue
            lib_info = self.libs.get_lib(function.lib_id)
            if filter_lib(lib_info.name):
                addr2line.add_addr(lib_info.name, lib_info.build_id,
                                   function.start_addr, function.start_addr)
                addr2line.add_addr(lib_info.name, lib_info.build_id, function.start_addr,
                                   function.start_addr + function.addr_len - 1)
        # Request line for each addr in FunctionScope.addr_hit_map.
        for event in self.events.values():
            for lib in event.libraries:
                lib_info = self.libs.get_lib(lib.lib_id)
                if filter_lib(lib_info.name):
                    for function in lib.functions.values():
                        func_addr = self.functions.id_to_func[function.func_id].start_addr
                        for addr in function.addr_hit_map:
                            addr2line.add_addr(lib_info.name, lib_info.build_id, func_addr, addr)
        addr2line.convert_addrs_to_lines(jobs)

        # Set line range for each function.
        for function in self.functions.id_to_func.values():
            if function.func_name == 'unknown':
                continue
            dso = addr2line.get_dso(self.libs.get_lib(function.lib_id).name)
            if not dso:
                continue
            start_source = addr2line.get_addr_source(dso, function.start_addr)
            end_source = addr2line.get_addr_source(dso, function.start_addr + function.addr_len - 1)
            if not start_source or not end_source:
                continue
            start_file_path, start_line = start_source[-1]
            end_file_path, end_line = end_source[-1]
            if start_file_path != end_file_path or start_line > end_line:
                continue
            source_file = self.source_files.get_source_file(start_file_path)
            source_file.request_lines(start_line, end_line)
            function.source_info = (source_file.file_id, start_line, end_line)

        # Build FunctionScope.line_hit_map.
        for event in self.events.values():
            for lib in event.libraries:
                dso = addr2line.get_dso(self.libs.get_lib(lib.lib_id).name)
                if not dso:
                    continue
                for function in lib.functions.values():
                    for addr in function.addr_hit_map:
                        source = addr2line.get_addr_source(dso, addr)
                        if not source:
                            continue
                        for file_path, line in source:
                            source_file = self.source_files.get_source_file(file_path)
                            # Show [line - 5, line + 5] of the line hit by a sample.
                            source_file.request_lines(line - 5, line + 5)
                            count_info = function.addr_hit_map[addr]
                            function.build_line_hit_map(source_file.file_id, line, count_info[0],
                                                        count_info[1])

        # Collect needed source code in SourceFileSet.
        self.source_files.load_source_code(source_dirs)

    def add_disassembly(self, filter_lib: Callable[[str], bool],
                        jobs: int, disassemble_job_size: int):
        """ Collect disassembly information:
            1. Use objdump to collect disassembly for each function in FunctionSet.
            2. Set flag to dump addr_hit_map when generating record info.
        """
        objdump = Objdump(self.ndk_path, self.binary_finder)
        lib_functions: Dict[int, List[Function]] = collections.defaultdict(list)

        for function in self.functions.id_to_func.values():
            if function.func_name == 'unknown':
                continue
            lib_functions[function.lib_id].append(function)

        with ThreadPoolExecutor(jobs) as executor:
            futures: List[Future] = []
            all_tasks = []
            for lib_id, functions in lib_functions.items():
                lib = self.libs.get_lib(lib_id)
                if not filter_lib(lib.name):
                    continue
                dso_info = objdump.get_dso_info(lib.name, lib.build_id)
                if not dso_info:
                    continue

                tasks = self.split_disassembly_jobs(functions, disassemble_job_size)
                logging.debug('create %d jobs to disassemble %d functions in %s',
                              len(tasks), len(functions), lib.name)
                for task in tasks:
                    futures.append(executor.submit(
                        self._disassemble_functions, objdump, dso_info, task))
                    all_tasks.append(task)

            for task, future in zip(all_tasks, futures):
                result = future.result()
                if result and len(result) == len(task):
                    for function, disassembly in zip(task, result):
                        function.disassembly = disassembly.lines

        logging.debug('finished all disassemble jobs')
        self.gen_addr_hit_map_in_record_info = True

    def split_disassembly_jobs(self, functions: List[Function],
                               disassemble_job_size: int) -> List[List[Function]]:
        """ Decide how to split the task of dissassembly functions in one library. """
        if not functions:
            return []
        functions.sort(key=lambda f: f.start_addr)
        result = []
        job_start_addr = None
        for function in functions:
            if (job_start_addr is None or
                    function.start_addr - job_start_addr > disassemble_job_size):
                job_start_addr = function.start_addr
                result.append([function])
            else:
                result[-1].append(function)
        return result

    def _disassemble_functions(self, objdump: Objdump, dso_info,
                               functions: List[Function]) -> Optional[List[Disassembly]]:
        addr_ranges = [AddrRange(f.start_addr, f.addr_len) for f in functions]
        return objdump.disassemble_functions(dso_info, addr_ranges)

    def gen_record_info(self) -> Dict[str, Any]:
        """ Return json data which will be used by report_html.js. """
        record_info = {}
        timestamp = self.meta_info.get('timestamp')
        if timestamp:
            t = datetime.datetime.fromtimestamp(int(timestamp))
        else:
            t = datetime.datetime.now()
        record_info['recordTime'] = t.strftime('%Y-%m-%d (%A) %H:%M:%S')

        product_props = self.meta_info.get('product_props')
        machine_type = self.arch
        if product_props:
            manufacturer, model, name = product_props.split(':')
            machine_type = '%s (%s) by %s, arch %s' % (model, name, manufacturer, self.arch)
        record_info['machineType'] = machine_type
        record_info['androidVersion'] = self.meta_info.get('android_version', '')
        record_info['androidBuildFingerprint'] = self.meta_info.get('android_build_fingerprint', '')
        record_info['kernelVersion'] = self.meta_info.get('kernel_version', '')
        record_info['recordCmdline'] = self.cmdline
        record_info['totalSamples'] = self.total_samples
        record_info['processNames'] = self._gen_process_names()
        record_info['threadNames'] = self._gen_thread_names()
        record_info['libList'] = self._gen_lib_list()
        record_info['functionMap'] = self._gen_function_map()
        record_info['sampleInfo'] = self._gen_sample_info()
        record_info['sourceFiles'] = self._gen_source_files()
        return record_info

    def _gen_process_names(self) -> Dict[int, str]:
        process_names: Dict[int, str] = {}
        for event in self.events.values():
            for process in event.processes.values():
                process_names[process.pid] = process.name
        return process_names

    def _gen_thread_names(self) -> Dict[int, str]:
        thread_names: Dict[int, str] = {}
        for event in self.events.values():
            for process in event.processes.values():
                for thread in process.threads.values():
                    thread_names[thread.tid] = thread.name
        return thread_names

    def _gen_lib_list(self) -> List[str]:
        return [modify_text_for_html(lib.name) for lib in self.libs.libs]

    def _gen_function_map(self) -> Dict[int, Any]:
        func_map: Dict[int, Any] = {}
        for func_id in sorted(self.functions.id_to_func):
            function = self.functions.id_to_func[func_id]
            func_data = {}
            func_data['l'] = function.lib_id
            func_data['f'] = modify_text_for_html(function.func_name)
            if function.source_info:
                func_data['s'] = function.source_info
            if function.disassembly:
                disassembly_list = []
                for code, addr in function.disassembly:
                    disassembly_list.append(
                        [modify_text_for_html(code),
                         hex_address_for_json(addr)])
                func_data['d'] = disassembly_list
            func_map[func_id] = func_data
        return func_map

    def _gen_sample_info(self) -> List[Dict[str, Any]]:
        return [event.get_sample_info(self.gen_addr_hit_map_in_record_info)
                for event in self.events.values()]

    def _gen_source_files(self) -> List[Dict[str, Any]]:
        source_files = sorted(self.source_files.path_to_source_files.values(),
                              key=lambda x: x.file_id)
        file_list = []
        for source_file in source_files:
            file_data = {}
            if not source_file.real_path:
                file_data['path'] = ''
                file_data['code'] = {}
            else:
                file_data['path'] = source_file.real_path
                code_map = {}
                for line in source_file.line_to_code:
                    code_map[line] = modify_text_for_html(source_file.line_to_code[line])
                file_data['code'] = code_map
            file_list.append(file_data)
        return file_list


URLS = {
    'jquery': 'https://ajax.googleapis.com/ajax/libs/jquery/3.3.1/jquery.min.js',
    'bootstrap4-css': 'https://stackpath.bootstrapcdn.com/bootstrap/4.1.2/css/bootstrap.min.css',
    'bootstrap4-popper':
        'https://cdnjs.cloudflare.com/ajax/libs/popper.js/1.12.9/umd/popper.min.js',
    'bootstrap4': 'https://stackpath.bootstrapcdn.com/bootstrap/4.1.2/js/bootstrap.min.js',
    'dataTable': 'https://cdn.datatables.net/1.10.19/js/jquery.dataTables.min.js',
    'dataTable-bootstrap4': 'https://cdn.datatables.net/1.10.19/js/dataTables.bootstrap4.min.js',
    'dataTable-css': 'https://cdn.datatables.net/1.10.19/css/dataTables.bootstrap4.min.css',
    'gstatic-charts': 'https://www.gstatic.com/charts/loader.js',
}


class ReportGenerator(object):

    def __init__(self, html_path: Union[Path, str]):
        self.hw = HtmlWriter(html_path)
        self.hw.open_tag('html')
        self.hw.open_tag('head')
        for css in ['bootstrap4-css', 'dataTable-css']:
            self.hw.open_tag('link', rel='stylesheet', type='text/css', href=URLS[css]).close_tag()
        for js in ['jquery', 'bootstrap4-popper', 'bootstrap4', 'dataTable', 'dataTable-bootstrap4',
                   'gstatic-charts']:
            self.hw.open_tag('script', src=URLS[js]).close_tag()

        self.hw.open_tag('script').add(
            "google.charts.load('current', {'packages': ['corechart', 'table']});").close_tag()
        self.hw.open_tag('style', type='text/css').add("""
            .colForLine { width: 50px; text-align: right; }
            .colForCount { width: 100px; text-align: right; }
            .tableCell { font-size: 17px; }
            .boldTableCell { font-weight: bold; font-size: 17px; }
            .textRight { text-align: right; }
            """).close_tag()
        self.hw.close_tag('head')
        self.hw.open_tag('body')

    def write_content_div(self):
        self.hw.open_tag('div', id='report_content').close_tag()

    def write_record_data(self, record_data: Dict[str, Any]):
        self.hw.open_tag('script', id='record_data', type='application/json')
        self.hw.add(json.dumps(record_data))
        self.hw.close_tag()

    def write_script(self):
        self.hw.open_tag('script').add_file('report_html.js').close_tag()

    def finish(self):
        self.hw.close_tag('body')
        self.hw.close_tag('html')
        self.hw.close()


def get_args() -> argparse.Namespace:
    parser = BaseArgumentParser(description='report profiling data')
    parser.add_argument('-i', '--record_file', nargs='+', default=['perf.data'], help="""
                        Set profiling data file to report.""")
    parser.add_argument('-o', '--report_path', default='report.html', help='Set output html file')
    parser.add_argument('--min_func_percent', default=0.01, type=float, help="""
                        Set min percentage of functions shown in the report.
                        For example, when set to 0.01, only functions taking >= 0.01%% of total
                        event count are collected in the report.""")
    parser.add_argument('--min_callchain_percent', default=0.01, type=float, help="""
                        Set min percentage of callchains shown in the report.
                        It is used to limit nodes shown in the function flamegraph. For example,
                        when set to 0.01, only callchains taking >= 0.01%% of the event count of
                        the starting function are collected in the report.""")
    parser.add_argument('--add_source_code', action='store_true', help='Add source code.')
    parser.add_argument('--source_dirs', nargs='+', help='Source code directories.')
    parser.add_argument('--add_disassembly', action='store_true', help='Add disassembled code.')
    parser.add_argument('--disassemble-job-size', type=int, default=1024*1024,
                        help='address range for one disassemble job')
    parser.add_argument('--binary_filter', nargs='+', help="""Annotate source code and disassembly
                        only for selected binaries, whose recorded paths contains [BINARY_FILTER] as
                        a substring. Example: to select binaries belonging to an app with package
                        name 'com.example.myapp', use `--binary_filter com.example.myapp`.
                        """)
    parser.add_argument(
        '-j', '--jobs', type=int, default=os.cpu_count(),
        help='Use multithreading to speed up disassembly and source code annotation.')
    parser.add_argument('--ndk_path', nargs=1, help='Find tools in the ndk path.')
    parser.add_argument('--no_browser', action='store_true', help="Don't open report in browser.")
    parser.add_argument('--aggregate-by-thread-name', action='store_true', help="""aggregate
                        samples by thread name instead of thread id. This is useful for
                        showing multiple perf.data generated for the same app.""")
    parser.add_report_lib_options()
    return parser.parse_args()


def main():
    sys.setrecursionlimit(MAX_CALLSTACK_LENGTH * 2 + 50)
    args = get_args()

    # 1. Process args.
    binary_cache_path = 'binary_cache'
    if not os.path.isdir(binary_cache_path):
        if args.add_source_code or args.add_disassembly:
            log_exit("""binary_cache/ doesn't exist. Can't add source code or disassembled code
                        without collected binaries. Please run binary_cache_builder.py to
                        collect binaries for current profiling data, or run app_profiler.py
                        without -nb option.""")
        binary_cache_path = None

    if args.add_source_code and not args.source_dirs:
        log_exit('--source_dirs is needed to add source code.')
    build_addr_hit_map = args.add_source_code or args.add_disassembly
    ndk_path = None if not args.ndk_path else args.ndk_path[0]
    if args.jobs < 1:
        log_exit('Invalid --jobs option.')

    # 2. Produce record data.
    record_data = RecordData(binary_cache_path, ndk_path, build_addr_hit_map)
    for record_file in args.record_file:
        record_data.load_record_file(record_file, args.report_lib_options)
    if args.aggregate_by_thread_name:
        record_data.aggregate_by_thread_name()
    record_data.limit_percents(args.min_func_percent, args.min_callchain_percent)
    record_data.sort_call_graph_by_function_name()

    def filter_lib(lib_name: str) -> bool:
        if not args.binary_filter:
            return True
        for binary in args.binary_filter:
            if binary in lib_name:
                return True
        return False
    if args.add_source_code:
        record_data.add_source_code(args.source_dirs, filter_lib, args.jobs)
    if args.add_disassembly:
        record_data.add_disassembly(filter_lib, args.jobs, args.disassemble_job_size)

    # 3. Generate report html.
    report_generator = ReportGenerator(args.report_path)
    report_generator.write_script()
    report_generator.write_content_div()
    report_generator.write_record_data(record_data.gen_record_info())
    report_generator.finish()

    if not args.no_browser:
        open_report_in_browser(args.report_path)
    logging.info("Report generated at '%s'." % args.report_path)


if __name__ == '__main__':
    main()