#!/usr/bin/env python3
# vim: set syntax=python ts=4 :
# Copyright (c) 2020 Intel Corporation
# SPDX-License-Identifier: Apache-2.0
"""Zephyr Test Runner (twister)

Also check the "User and Developer Guides" at https://docs.zephyrproject.org/

This script scans for the set of unit test applications in the git
repository and attempts to execute them. By default, it tries to
build each test case on one platform per architecture, using a precedence
list defined in an architecture configuration file, and if possible
run the tests in any available emulators or simulators on the system.

Test cases are detected by the presence of a 'testcase.yaml' or a sample.yaml
files in the application's project directory. This file may contain one or more
blocks, each identifying a test scenario. The title of the block is a name for
the test case, which only needs to be unique for the test cases specified in
that testcase meta-data. The full canonical name for each test case is <path to
test case>/<block>.

Each test block in the testcase meta data can define the following key/value
pairs:

  tags: <list of tags> (required)
    A set of string tags for the testcase. Usually pertains to
    functional domains but can be anything. Command line invocations
    of this script can filter the set of tests to run based on tag.

  skip: <True|False> (default False)
    skip testcase unconditionally. This can be used for broken tests.

  slow: <True|False> (default False)
    Don't build or run this test case unless --enable-slow was passed
    in on the command line. Intended for time-consuming test cases
    that are only run under certain circumstances, like daily
    builds.

  extra_args: <list of extra arguments>
    Extra cache entries to pass to CMake when building or running the
    test case.

  extra_configs: <list of extra configurations>
    Extra configuration options to be merged with a master prj.conf
    when building or running the test case.

  build_only: <True|False> (default False)
    If true, don't try to run the test even if the selected platform
    supports it.

  build_on_all: <True|False> (default False)
    If true, attempt to build test on all available platforms.

  depends_on: <list of features>
    A board or platform can announce what features it supports, this option
    will enable the test only those platforms that provide this feature.

  min_ram: <integer>
    minimum amount of RAM needed for this test to build and run. This is
    compared with information provided by the board metadata.

  min_flash: <integer>
    minimum amount of ROM needed for this test to build and run. This is
    compared with information provided by the board metadata.

  timeout: <number of seconds>
    Length of time to run test in emulator before automatically killing it.
    Default to 60 seconds.

  arch_allow: <list of arches, such as x86, arm, arc>
    Set of architectures that this test case should only be run for.

  arch_exclude: <list of arches, such as x86, arm, arc>
    Set of architectures that this test case should not run on.

  platform_allow: <list of platforms>
    Set of platforms that this test case should only be run for.

  platform_exclude: <list of platforms>
    Set of platforms that this test case should not run on.

  extra_sections: <list of extra binary sections>
    When computing sizes, twister will report errors if it finds
    extra, unexpected sections in the Zephyr binary unless they are named
    here. They will not be included in the size calculation.

  filter: <expression>
    Filter whether the testcase should be run by evaluating an expression
    against an environment containing the following values:

    { ARCH : <architecture>,
      PLATFORM : <platform>,
      <all CONFIG_* key/value pairs in the test's generated defconfig>,
      <all DT_* key/value pairs in the test's generated device tree file>,
      <all CMake key/value pairs in the test's generated CMakeCache.txt file>,
      *<env>: any environment variable available
    }

    The grammar for the expression language is as follows:

    expression ::= expression "and" expression
                 | expression "or" expression
                 | "not" expression
                 | "(" expression ")"
                 | symbol "==" constant
                 | symbol "!=" constant
                 | symbol "<" number
                 | symbol ">" number
                 | symbol ">=" number
                 | symbol "<=" number
                 | symbol "in" list
                 | symbol ":" string
                 | symbol

    list ::= "[" list_contents "]"

    list_contents ::= constant
                    | list_contents "," constant

    constant ::= number
               | string


    For the case where expression ::= symbol, it evaluates to true
    if the symbol is defined to a non-empty string.

    Operator precedence, starting from lowest to highest:

        or (left associative)
        and (left associative)
        not (right associative)
        all comparison operators (non-associative)

    arch_allow, arch_exclude, platform_allow, platform_exclude
    are all syntactic sugar for these expressions. For instance

        arch_exclude = x86 arc

    Is the same as:

        filter = not ARCH in ["x86", "arc"]

    The ':' operator compiles the string argument as a regular expression,
    and then returns a true value only if the symbol's value in the environment
    matches. For example, if CONFIG_SOC="stm32f107xc" then

        filter = CONFIG_SOC : "stm.*"

    Would match it.

The set of test cases that actually run depends on directives in the testcase
filed and options passed in on the command line. If there is any confusion,
running with -v or examining the discard report (twister_discard.csv)
can help show why particular test cases were skipped.

Metrics (such as pass/fail state and binary size) for the last code
release are stored in scripts/release/twister_last_release.csv.
To update this, pass the --all --release options.

To load arguments from a file, write '+' before the file name, e.g.,
+file_name. File content must be one or more valid arguments separated by
line break instead of white spaces.

Most everyday users will run with no arguments.

"""

import os
import argparse
import sys
import logging
import time
import itertools
import shutil
from collections import OrderedDict
import multiprocessing
from itertools import islice
import csv
from colorama import Fore
from pathlib import Path
from multiprocessing.managers import BaseManager
import queue

ZEPHYR_BASE = os.getenv("ZEPHYR_BASE")
if not ZEPHYR_BASE:
    # This file has been zephyr/scripts/twister for years,
    # and that is not going to change anytime soon. Let the user
    # run this script as ./scripts/twister without making them
    # set ZEPHYR_BASE.
    ZEPHYR_BASE = str(Path(__file__).resolve().parents[1])

    # Propagate this decision to child processes.
    os.environ['ZEPHYR_BASE'] = ZEPHYR_BASE

    print(f'ZEPHYR_BASE unset, using "{ZEPHYR_BASE}"')

try:
    from anytree import RenderTree, Node, find
except ImportError:
    print("Install the anytree module to use the --test-tree option")

try:
    from tabulate import tabulate
except ImportError:
    print("Install tabulate python module with pip to use --device-testing option.")

sys.path.insert(0, os.path.join(ZEPHYR_BASE, "scripts/pylib/twister"))

from twisterlib import HardwareMap, TestSuite, SizeCalculator, CoverageTool, ExecutionCounter

logger = logging.getLogger('twister')
logger.setLevel(logging.DEBUG)

def size_report(sc):
    logger.info(sc.filename)
    logger.info("SECTION NAME             VMA        LMA     SIZE  HEX SZ TYPE")
    for v in sc.sections:
        logger.info("%-17s 0x%08x 0x%08x %8d 0x%05x %-7s" %
                    (v["name"], v["virt_addr"], v["load_addr"], v["size"], v["size"],
                     v["type"]))

    logger.info("Totals: %d bytes (ROM), %d bytes (RAM)" %
                (sc.rom_size, sc.ram_size))
    logger.info("")


def export_tests(filename, tests):
    with open(filename, "wt") as csvfile:
        fieldnames = ['section', 'subsection', 'title', 'reference']
        cw = csv.DictWriter(csvfile, fieldnames, lineterminator=os.linesep)
        for test in tests:
            data = test.split(".")
            if len(data) > 1:
                subsec = " ".join(data[1].split("_")).title()
                rowdict = {
                    "section": data[0].capitalize(),
                    "subsection": subsec,
                    "title": test,
                    "reference": test
                }
                cw.writerow(rowdict)
            else:
                logger.error("{} can't be exported: ".format(test))


def parse_arguments():
    parser = argparse.ArgumentParser(
        description=__doc__,
        formatter_class=argparse.RawDescriptionHelpFormatter)
    parser.fromfile_prefix_chars = "+"

    case_select = parser.add_argument_group("Test case selection",
                                            """
Artificially long but functional example:
    $ ./scripts/twister -v     \\
      --testcase-root tests/ztest/base    \\
      --testcase-root tests/kernel   \\
      --test      tests/ztest/base/testing.ztest.verbose_0  \\
      --test      tests/kernel/fifo/fifo_api/kernel.fifo

   "kernel.fifo.poll" is one of the test section names in
                                 __/fifo_api/testcase.yaml
    """)

    compare_group_option = parser.add_mutually_exclusive_group()

    platform_group_option = parser.add_mutually_exclusive_group()

    run_group_option = parser.add_mutually_exclusive_group()

    serial = parser.add_mutually_exclusive_group(required="--device-testing" in sys.argv)

    test_or_build = parser.add_mutually_exclusive_group()

    test_xor_subtest = case_select.add_mutually_exclusive_group()

    valgrind_asan_group = parser.add_mutually_exclusive_group()

    case_select.add_argument(
        "-E",
        "--save-tests",
        metavar="FILENAME",
        action="store",
        help="Append list of tests and platforms to be run to file.")

    case_select.add_argument(
        "-F",
        "--load-tests",
        metavar="FILENAME",
        action="store",
        help="Load list of tests and platforms to be run from file.")

    case_select.add_argument(
        "-T", "--testcase-root", action="append", default=[],
        help="Base directory to recursively search for test cases. All "
             "testcase.yaml files under here will be processed. May be "
             "called multiple times. Defaults to the 'samples/' and "
             "'tests/' directories at the base of the Zephyr tree.")

    case_select.add_argument(
        "-f",
        "--only-failed",
        action="store_true",
        help="Run only those tests that failed the previous twister run "
             "invocation.")

    case_select.add_argument("--list-tests", action="store_true",
                             help="""List of all sub-test functions recursively found in
        all --testcase-root arguments. Note different sub-tests can share
        the same section name and come from different directories.
        The output is flattened and reports --sub-test names only,
        not their directories. For instance net.socket.getaddrinfo_ok
        and net.socket.fd_set belong to different directories.
        """)

    case_select.add_argument("--list-test-duplicates", action="store_true",
                             help="""List tests with duplicate identifiers.
        """)

    case_select.add_argument("--test-tree", action="store_true",
                             help="""Output the testsuite in a tree form""")

    compare_group_option.add_argument("--compare-report",
                        help="Use this report file for size comparison")

    compare_group_option.add_argument(
        "-m", "--last-metrics", action="store_true",
        help="Instead of comparing metrics from the last --release, "
             "compare with the results of the previous twister "
             "invocation")

    platform_group_option.add_argument(
        "-G",
        "--integration",
        action="store_true",
        help="Run integration tests")

    platform_group_option.add_argument(
        "--emulation-only", action="store_true",
        help="Only build and run emulation platforms")

    run_group_option.add_argument(
        "--device-testing", action="store_true",
        help="Test on device directly. Specify the serial device to "
             "use with the --device-serial option.")

    run_group_option.add_argument("--generate-hardware-map",
                        help="""Probe serial devices connected to this platform
                        and create a hardware map file to be used with
                        --device-testing
                        """)

    serial.add_argument("--device-serial",
                        help="""Serial device for accessing the board
                        (e.g., /dev/ttyACM0)
                        """)

    serial.add_argument("--device-serial-pty",
                        help="""Script for controlling pseudoterminal.
                        Twister believes that it interacts with a terminal
                        when it actually interacts with the script.

                        E.g "twister --device-testing
                        --device-serial-pty <script>
                        """)

    serial.add_argument("--hardware-map",
                        help="""Load hardware map from a file. This will be used
                        for testing on hardware that is listed in the file.
                        """)

    test_or_build.add_argument(
        "-b", "--build-only", action="store_true",
        help="Only build the code, do not execute any of it in QEMU")

    test_or_build.add_argument(
        "--test-only", action="store_true",
        help="""Only run device tests with current artifacts, do not build
             the code""")

    test_xor_subtest.add_argument(
        "-s", "--test", action="append",
        help="Run only the specified test cases. These are named by "
             "<path/relative/to/Zephyr/base/section.name.in.testcase.yaml>")

    test_xor_subtest.add_argument(
        "--sub-test", action="append",
        help="""Recursively find sub-test functions and run the entire
        test section where they were found, including all sibling test
        functions. Sub-tests are named by:
        section.name.in.testcase.yaml.function_name_without_test_prefix
        Example: In kernel.fifo.fifo_loop: 'kernel.fifo' is a section name
        and 'fifo_loop' is a name of a function found in main.c without test prefix.
        """)

    valgrind_asan_group.add_argument(
        "--enable-valgrind", action="store_true",
        help="""Run binary through valgrind and check for several memory access
        errors. Valgrind needs to be installed on the host. This option only
        works with host binaries such as those generated for the native_posix
        configuration and is mutual exclusive with --enable-asan.
        """)

    valgrind_asan_group.add_argument(
        "--enable-asan", action="store_true",
        help="""Enable address sanitizer to check for several memory access
        errors. Libasan needs to be installed on the host. This option only
        works with host binaries such as those generated for the native_posix
        configuration and is mutual exclusive with --enable-valgrind.
        """)

    # Start of individual args place them in alpha-beta order

    board_root_list = ["%s/boards" % ZEPHYR_BASE,
                       "%s/scripts/pylib/twister/boards" % ZEPHYR_BASE]

    parser.add_argument(
        "-A", "--board-root", action="append", default=board_root_list,
        help="""Directory to search for board configuration files. All .yaml
files in the directory will be processed. The directory should have the same
structure in the main Zephyr tree: boards/<arch>/<board_name>/""")

    parser.add_argument(
        "-a", "--arch", action="append",
        help="Arch filter for testing. Takes precedence over --platform. "
             "If unspecified, test all arches. Multiple invocations "
             "are treated as a logical 'or' relationship")

    parser.add_argument(
        "-B", "--subset",
        help="Only run a subset of the tests, 1/4 for running the first 25%%, "
             "3/5 means run the 3rd fifth of the total. "
             "This option is useful when running a large number of tests on "
             "different hosts to speed up execution time.")

    parser.add_argument("-C", "--coverage", action="store_true",
                        help="Generate coverage reports. Implies "
                             "--enable-coverage.")

    parser.add_argument(
        "-c", "--clobber-output", action="store_true",
        help="Cleaning the output directory will simply delete it instead "
             "of the default policy of renaming.")

    parser.add_argument(
        "--cmake-only", action="store_true",
        help="Only run cmake, do not build or run.")

    parser.add_argument("--coverage-basedir", default=ZEPHYR_BASE,
                        help="Base source directory for coverage report.")

    parser.add_argument("--coverage-platform", action="append", default=[],
                        help="Platforms to run coverage reports on. "
                             "This option may be used multiple times. "
                             "Default to what was selected with --platform.")

    parser.add_argument("--coverage-tool", choices=['lcov', 'gcovr'], default='lcov',
                        help="Tool to use to generate coverage report.")

    parser.add_argument(
        "-D", "--all-deltas", action="store_true",
        help="Show all footprint deltas, positive or negative. Implies "
             "--footprint-threshold=0")

    parser.add_argument(
        "--device-serial-baud", action="store", default=None,
        help="Serial device baud rate (default 115200)")

    parser.add_argument("--disable-asserts", action="store_false",
                        dest="enable_asserts",
                        help="deprecated, left for compatibility")

    parser.add_argument(
        "--disable-unrecognized-section-test", action="store_true",
        default=False,
        help="Skip the 'unrecognized section' test.")

    parser.add_argument("-e", "--exclude-tag", action="append",
                        help="Specify tags of tests that should not run. "
                             "Default is to run all tests with all tags.")

    parser.add_argument("--enable-coverage", action="store_true",
                        help="Enable code coverage using gcov.")

    parser.add_argument(
        "--enable-lsan", action="store_true",
        help="""Enable leak sanitizer to check for heap memory leaks.
        Libasan needs to be installed on the host. This option only
        works with host binaries such as those generated for the native_posix
        configuration and when --enable-asan is given.
        """)

    parser.add_argument(
        "--enable-ubsan", action="store_true",
        help="""Enable undefined behavior sanitizer to check for undefined
        behaviour during program execution. It uses an optional runtime library
        to provide better error diagnostics. This option only works with host
        binaries such as those generated for the native_posix configuration.
        """)

    parser.add_argument("--enable-size-report", action="store_true",
                        help="Enable expensive computation of RAM/ROM segment sizes.")

    parser.add_argument("--export-tests", action="store",
                        metavar="FILENAME",
                        help="Export tests case meta-data to a file in CSV format."
                             "Test instances can be exported per target by supplying "
                             "the platform name using --platform option. (tests for only "
                             " one platform can be exported at a time)")

    parser.add_argument(
        "--filter", choices=['buildable', 'runnable'],
        default='buildable',
        help="""Filter tests to be built and executed. By default everything is
        built and if a test is runnable (emulation or a connected device), it
        is run. This option allows for example to only build tests that can
        actually be run. Runnable is a subset of buildable.""")

    parser.add_argument("--force-toolchain", action="store_true",
                        help="Do not filter based on toolchain, use the set "
                             " toolchain unconditionally")

    parser.add_argument("--gcov-tool", default=None,
                        help="Path to the gcov tool to use for code coverage "
                             "reports")

    parser.add_argument(
        "-H", "--footprint-threshold", type=float, default=5,
        help="When checking test case footprint sizes, warn the user if "
             "the new app size is greater then the specified percentage "
             "from the last release. Default is 5. 0 to warn on any "
             "increase on app size.")

    parser.add_argument(
        "-i", "--inline-logs", action="store_true",
        help="Upon test failure, print relevant log data to stdout "
             "instead of just a path to it.")

    parser.add_argument(
        "-j", "--jobs", type=int,
        help="Number of jobs for building, defaults to number of CPU threads, "
             "overcommited by factor 2 when --build-only.")

    parser.add_argument(
        "--json-report", action="store_true",
        help="""Generate a JSON file with all test results. [Experimental]
        """)

    parser.add_argument(
        "-K", "--force-platform", action="store_true",
        help="""Force testing on selected platforms,
        even if they are excluded in the test configuration (testcase.yaml)."""
    )

    parser.add_argument(
        "-l", "--all", action="store_true",
        help="Build/test on all platforms. Any --platform arguments "
             "ignored.")

    parser.add_argument("--list-tags", action="store_true",
                        help="List all tags occurring in selected tests.")

    parser.add_argument("--log-file", metavar="FILENAME", action="store",
                        help="Specify a file where to save logs.")

    parser.add_argument(
        "-M", "--runtime-artifact-cleanup", action="store_true",
        help="Delete artifacts of passing tests.")

    parser.add_argument(
        "-N", "--ninja", action="store_true",
        help="Use the Ninja generator with CMake.")

    parser.add_argument(
        "-n", "--no-clean", action="store_true",
        help="Re-use the outdir before building. Will result in "
             "faster compilation since builds will be incremental.")

    parser.add_argument(
        "-O", "--outdir",
        default=os.path.join(os.getcwd(), "twister-out"),
        help="Output directory for logs and binaries. "
             "Default is 'twister-out' in the current directory. "
             "This directory will be cleaned unless '--no-clean' is set. "
             "The '--clobber-output' option controls what cleaning does.")

    parser.add_argument(
        "-o", "--report-dir",
        help="""Output reports containing results of the test run into the
        specified directory.
        The output will be both in CSV and JUNIT format
        (twister.csv and twister.xml).
        """)

    parser.add_argument("--overflow-as-errors", action="store_true",
                        help="Treat RAM/SRAM overflows as errors.")


    parser.add_argument("-P", "--exclude-platform", action="append", default=[],
            help="""Exclude platforms and do not build or run any tests
            on those platforms. This option can be called multiple times.
            """
            )

    parser.add_argument("--persistent-hardware-map", action='store_true',
                        help="""With --generate-hardware-map, tries to use
                        persistent names for serial devices on platforms
                        that support this feature (currently only Linux).
                        """)

    parser.add_argument(
        "-p", "--platform", action="append",
        help="Platform filter for testing. This option may be used multiple "
             "times. Testcases will only be built/run on the platforms "
             "specified. If this option is not used, then platforms marked "
             "as default in the platform metadata file will be chosen "
             "to build and test. ")

    parser.add_argument(
        "--platform-reports", action="store_true",
        help="""Create individual reports for each platform.
        """)

    parser.add_argument("--pre-script",
                        help="""specify a pre script. This will be executed
                        before device handler open serial port and invoke runner.
                        """)

    parser.add_argument("-Q", "--error-on-deprecations", action="store_false",
                        help="Error on deprecation warnings.")

    parser.add_argument(
        "--quarantine-list",
        metavar="FILENAME",
        help="Load list of test scenarios under quarantine. The entries in "
             "the file need to correspond to the test scenarios names as in "
             "corresponding tests .yaml files. These scenarios "
             "will be skipped with quarantine as the reason.")

    parser.add_argument(
        "--quarantine-verify",
        action="store_true",
        help="Use the list of test scenarios under quarantine and run them"
             "to verify their current status.")

    parser.add_argument("-R", "--enable-asserts", action="store_true",
                        default=True,
                        help="deprecated, left for compatibility")

    parser.add_argument(
        "-r", "--release", action="store_true",
        help="Update the benchmark database with the results of this test "
             "run. Intended to be run by CI when tagging an official "
             "release. This database is used as a basis for comparison "
             "when looking for deltas in metrics such as footprint.")

    parser.add_argument("--report-excluded",
                        action="store_true",
                        help="""List all tests that are never run based on current scope and
            coverage. If you are looking for accurate results, run this with
            --all, but this will take a while...""")

    parser.add_argument(
        "--report-name",
        help="""Create a report with a custom name.
        """)

    parser.add_argument(
        "--report-suffix",
        help="""Add a suffix to all generated file names, for example to add a
        version or a commit ID.
        """)

    parser.add_argument(
        "--retry-failed", type=int, default=0,
        help="Retry failing tests again, up to the number of times specified.")

    parser.add_argument(
        "--retry-interval", type=int, default=60,
        help="Retry failing tests after specified period of time.")

    parser.add_argument(
        "-S", "--enable-slow", action="store_true",
        help="Execute time-consuming test cases that have been marked "
             "as 'slow' in testcase.yaml. Normally these are only built.")

    parser.add_argument(
        "--show-footprint", action="store_true",
        help="Show footprint statistics and deltas since last release."
    )

    parser.add_argument(
        "-t", "--tag", action="append",
        help="Specify tags to restrict which tests to run by tag value. "
             "Default is to not do any tag filtering. Multiple invocations "
             "are treated as a logical 'or' relationship.")

    parser.add_argument("--timestamps",
                        action="store_true",
                        help="Print all messages with time stamps.")

    parser.add_argument(
        "-u",
        "--no-update",
        action="store_true",
         help="Do not update the results of the last run of twister.")

    parser.add_argument(
        "-v",
        "--verbose",
        action="count",
        default=0,
        help="Emit debugging information, call multiple times to increase "
             "verbosity.")

    parser.add_argument("-W", "--disable-warnings-as-errors", action="store_true",
                        help="Do not treat warning conditions as errors.")

    parser.add_argument(
        "--west-flash", nargs='?', const=[],
        help="""Uses west instead of ninja or make to flash when running with
             --device-testing. Supports comma-separated argument list.

        E.g "twister --device-testing --device-serial /dev/ttyACM0
                         --west-flash="--board-id=foobar,--erase"
        will translate to "west flash -- --board-id=foobar --erase"

        NOTE: device-testing must be enabled to use this option.
        """
    )
    parser.add_argument(
        "--west-runner",
        help="""Uses the specified west runner instead of default when running
             with --west-flash.

        E.g "twister --device-testing --device-serial /dev/ttyACM0
                         --west-flash --west-runner=pyocd"
        will translate to "west flash --runner pyocd"

        NOTE: west-flash must be enabled to use this option.
        """
    )

    parser.add_argument(
        "-X", "--fixture", action="append", default=[],
        help="Specify a fixture that a board might support.")

    parser.add_argument(
        "-x", "--extra-args", action="append", default=[],
        help="""Extra CMake cache entries to define when building test cases.
        May be called multiple times. The key-value entries will be
        prefixed with -D before being passed to CMake.
        E.g
        "twister -x=USE_CCACHE=0"
        will translate to
        "cmake -DUSE_CCACHE=0"
        which will ultimately disable ccache.
        """
    )

    parser.add_argument(
        "-y", "--dry-run", action="store_true",
        help="""Create the filtered list of test cases, but don't actually
        run them. Useful if you're just interested in the discard report
        generated for every run and saved in the specified output
        directory (twister_discard.csv).
        """)

    parser.add_argument(
        "-z", "--size", action="append",
        help="Don't run twister. Instead, produce a report to "
             "stdout detailing RAM/ROM sizes on the specified filenames. "
             "All other command line arguments ignored.")

    return parser.parse_args()


def main():
    start_time = time.time()

    options = parse_arguments()
    previous_results = None
    # Cleanup
    if options.no_clean or options.only_failed or options.test_only:
        if os.path.exists(options.outdir):
            print("Keeping artifacts untouched")
    elif options.last_metrics:
        ls = os.path.join(options.outdir, "twister.csv")
        if os.path.exists(ls):
            with open(ls, "r") as fp:
                previous_results = fp.read()
        else:
            sys.exit(f"Can't compare metrics with non existing file {ls}")
    elif os.path.exists(options.outdir):
        if options.clobber_output:
            print("Deleting output directory {}".format(options.outdir))
            shutil.rmtree(options.outdir)
        else:
            for i in range(1, 100):
                new_out = options.outdir + ".{}".format(i)
                if not os.path.exists(new_out):
                    print("Renaming output directory to {}".format(new_out))
                    shutil.move(options.outdir, new_out)
                    break

    previous_results_file = None
    os.makedirs(options.outdir, exist_ok=True)
    if options.last_metrics and previous_results:
        previous_results_file = os.path.join(options.outdir, "baseline.csv")
        with open(previous_results_file, "w") as fp:
            fp.write(previous_results)

    # create file handler which logs even debug messages
    if options.log_file:
        fh = logging.FileHandler(options.log_file)
    else:
        fh = logging.FileHandler(os.path.join(options.outdir, "twister.log"))

    fh.setLevel(logging.DEBUG)

    # create console handler with a higher log level
    ch = logging.StreamHandler()

    VERBOSE = options.verbose
    if VERBOSE > 1:
        ch.setLevel(logging.DEBUG)
    else:
        ch.setLevel(logging.INFO)

    # create formatter and add it to the handlers
    if options.timestamps:
        formatter = logging.Formatter('%(asctime)s - %(levelname)s - %(message)s')
    else:
        formatter = logging.Formatter('%(levelname)-7s - %(message)s')

    formatter_file = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
    ch.setFormatter(formatter)
    fh.setFormatter(formatter_file)

    # add the handlers to logger
    logger.addHandler(ch)
    logger.addHandler(fh)

    hwm = HardwareMap()
    if options.generate_hardware_map:
        hwm.scan(persistent=options.persistent_hardware_map)
        hwm.save(options.generate_hardware_map)
        return

    if not options.device_testing and options.hardware_map:
        hwm.load(options.hardware_map)
        logger.info("Available devices:")
        table = []
        hwm.dump(connected_only=True)
        return

    if options.west_runner and options.west_flash is None:
        logger.error("west-runner requires west-flash to be enabled")
        sys.exit(1)

    if options.west_flash and not options.device_testing:
        logger.error("west-flash requires device-testing to be enabled")
        sys.exit(1)

    if options.coverage:
        options.enable_coverage = True

    if not options.coverage_platform:
        options.coverage_platform = options.platform

    if options.device_serial_pty and os.name == "nt":  # OS is Windows
        logger.error("--device-serial-pty is not supported on Windows OS")
        sys.exit(1)

    if options.size:
        for fn in options.size:
            size_report(SizeCalculator(fn, []))
        sys.exit(0)

    if options.subset:
        subset, sets = options.subset.split("/")
        if int(subset) > 0 and int(sets) >= int(subset):
            logger.info("Running only a subset: %s/%s" % (subset, sets))
        else:
            logger.error("You have provided a wrong subset value: %s." % options.subset)
            return

    if not options.testcase_root:
        options.testcase_root = [os.path.join(ZEPHYR_BASE, "tests"),
                                 os.path.join(ZEPHYR_BASE, "samples")]

    if options.show_footprint or options.compare_report or options.release:
        options.enable_size_report = True

    suite = TestSuite(options.board_root, options.testcase_root, options.outdir)

    # Check version of zephyr repo
    suite.check_zephyr_version()

    # Set testsuite options from command line.
    suite.build_only = options.build_only
    suite.cmake_only = options.cmake_only
    suite.cleanup = options.runtime_artifact_cleanup
    suite.test_only = options.test_only
    suite.enable_slow = options.enable_slow
    suite.device_testing = options.device_testing
    suite.fixtures = options.fixture
    suite.enable_asan = options.enable_asan
    suite.enable_lsan = options.enable_lsan
    suite.enable_ubsan = options.enable_ubsan
    suite.enable_coverage = options.enable_coverage
    suite.enable_valgrind = options.enable_valgrind
    suite.coverage_platform = options.coverage_platform
    suite.inline_logs = options.inline_logs
    suite.enable_size_report = options.enable_size_report
    suite.extra_args = options.extra_args
    suite.west_flash = options.west_flash
    suite.west_runner = options.west_runner
    suite.verbose = VERBOSE
    suite.warnings_as_errors = not options.disable_warnings_as_errors
    suite.integration = options.integration
    suite.overflow_as_errors = options.overflow_as_errors

    if options.ninja:
        suite.generator_cmd = "ninja"
        suite.generator = "Ninja"
    else:
        suite.generator_cmd = "make"
        suite.generator = "Unix Makefiles"

    # Set number of jobs
    if options.jobs:
        suite.jobs = options.jobs
    elif options.build_only:
        suite.jobs = multiprocessing.cpu_count() * 2
    else:
        suite.jobs = multiprocessing.cpu_count()
    logger.info("JOBS: %d" % suite.jobs)

    run_individual_tests = []

    if options.test:
        run_individual_tests = options.test

    num = suite.add_testcases(testcase_filter=run_individual_tests)
    if num == 0:
        logger.error("No test cases found at the specified location...")
        sys.exit(1)
    suite.add_configurations()

    if options.device_testing:
        if options.hardware_map:
            hwm.load(options.hardware_map)
            suite.duts = hwm.duts
            if not options.platform:
                options.platform = []
                for d in hwm.duts:
                    if d.connected:
                        options.platform.append(d.platform)

        elif options.device_serial or options.device_serial_pty:
            if options.platform and len(options.platform) == 1:
                if options.device_serial:
                    hwm.add_device(options.device_serial,
                                   options.platform[0],
                                   options.pre_script,
                                   False,
                                   baud=options.device_serial_baud
                                   )
                else:
                    hwm.add_device(options.device_serial_pty,
                                                 options.platform[0],
                                                 options.pre_script,
                                                 True)

                suite.duts = hwm.duts
            else:
                logger.error("""When --device-testing is used with
                             --device-serial or --device-serial-pty,
                             only one platform is allowed""")

    if suite.load_errors:
        sys.exit(1)

    if options.list_tags:
        tags = set()
        for _, tc in suite.testcases.items():
            tags = tags.union(tc.tags)

        for t in tags:
            print("- {}".format(t))

        return

    if not options.platform and (options.list_tests or options.test_tree or options.list_test_duplicates \
        or options.sub_test or options.export_tests):
        cnt = 0
        all_tests = suite.get_all_tests()

        if options.export_tests:
            export_tests(options.export_tests, all_tests)
            return

        if options.list_test_duplicates:
            import collections
            dupes = [item for item, count in collections.Counter(all_tests).items() if count > 1]
            if dupes:
                print("Tests with duplicate identifiers:")
                for dupe in dupes:
                    print("- {}".format(dupe))
                    for dc in suite.get_testcase(dupe):
                        print("  - {}".format(dc))
            else:
                print("No duplicates found.")
            return

        if options.sub_test:
            for st in options.sub_test:
                subtests = suite.get_testcase(st)
                for sti in subtests:
                    run_individual_tests.append(sti.name)

            if run_individual_tests:
                logger.info("Running the following tests:")
                for test in run_individual_tests:
                    print(" - {}".format(test))
            else:
                logger.info("Tests not found")
                return

        elif options.list_tests or options.test_tree:
            if options.test_tree:
                testsuite = Node("Testsuite")
                samples = Node("Samples", parent=testsuite)
                tests = Node("Tests", parent=testsuite)

            for test in sorted(all_tests):
                cnt = cnt + 1
                if options.list_tests:
                    print(" - {}".format(test))

                if options.test_tree:
                    if test.startswith("sample."):
                        sec = test.split(".")
                        area = find(samples, lambda node: node.name == sec[1] and node.parent == samples)
                        if not area:
                            area = Node(sec[1], parent=samples)

                        t = Node(test, parent=area)
                    else:
                        sec = test.split(".")
                        area = find(tests, lambda node: node.name == sec[0] and node.parent == tests)
                        if not area:
                            area = Node(sec[0], parent=tests)

                        if area and len(sec) > 2:
                            subarea = find(area, lambda node: node.name == sec[1] and node.parent == area)
                            if not subarea:
                                subarea = Node(sec[1], parent=area)
                            t = Node(test, parent=subarea)

            if options.list_tests:
                print("{} total.".format(cnt))

            if options.test_tree:
                for pre, _, node in RenderTree(testsuite):
                    print("%s%s" % (pre, node.name))
            return

    discards = []

    if options.report_suffix:
        last_run = os.path.join(options.outdir, "twister_{}.csv".format(options.report_suffix))
    else:
        last_run = os.path.join(options.outdir, "twister.csv")

    if options.quarantine_list:
        suite.load_quarantine(options.quarantine_list)

    if options.quarantine_verify:
        if not options.quarantine_list:
            logger.error("No quarantine list given to be verified")
            sys.exit(1)
        suite.quarantine_verify = options.quarantine_verify

    if options.only_failed:
        suite.load_from_file(last_run, filter_status=['skipped', 'passed'])
        suite.selected_platforms = set(p.platform.name for p in suite.instances.values())
    elif options.load_tests:
        suite.load_from_file(options.load_tests, filter_status=['skipped', 'error'])
        suite.selected_platforms = set(p.platform.name for p in suite.instances.values())
    elif options.test_only:
        # Get list of connected hardware and filter tests to only be run on connected hardware
        # in cases where no platform was specified when runn the tests.
        # If the platform does not exist in the hardware map, just skip it.
        connected_list = []
        if not options.platform:
            for connected in hwm.connected_hardware:
                if connected['connected']:
                    connected_list.append(connected['platform'])

        suite.load_from_file(last_run, filter_status=['skipped', 'error'],
                             filter_platform=connected_list)
        suite.selected_platforms = set(p.platform.name for p in suite.instances.values())
    else:
        discards = suite.apply_filters(
            enable_slow=options.enable_slow,
            platform=options.platform,
            exclude_platform=options.exclude_platform,
            arch=options.arch,
            tag=options.tag,
            exclude_tag=options.exclude_tag,
            force_toolchain=options.force_toolchain,
            all=options.all,
            emulation_only=options.emulation_only,
            run_individual_tests=run_individual_tests,
            runnable=(options.device_testing or options.filter == 'runnable'),
            force_platform=options.force_platform

        )

    if (options.export_tests or options.list_tests) and options.platform:
        if len(options.platform) > 1:
            logger.error("When exporting tests, only one platform "
                         "should be specified.")
            return

        for p in options.platform:
            inst = suite.get_platform_instances(p)
            if options.export_tests:
                tests = [x.testcase.cases for x in inst.values()]
                merged = list(itertools.chain(*tests))
                export_tests(options.export_tests, merged)
                return

            count = 0
            for i in inst.values():
                for c in i.testcase.cases:
                    print(f"- {c}")
                    count += 1

            print(f"Tests found: {count}")
        return

    if VERBOSE > 1 and discards:
        # if we are using command line platform filter, no need to list every
        # other platform as excluded, we know that already.
        # Show only the discards that apply to the selected platforms on the
        # command line

        for i, reason in discards.items():
            if options.platform and i.platform.name not in options.platform:
                continue
            logger.debug(
                "{:<25} {:<50} {}SKIPPED{}: {}".format(
                    i.platform.name,
                    i.testcase.name,
                    Fore.YELLOW,
                    Fore.RESET,
                    reason))

    if options.report_excluded:
        all_tests = suite.get_all_tests()
        to_be_run = set()
        for i, p in suite.instances.items():
            to_be_run.update(p.testcase.cases)

        if all_tests - to_be_run:
            print("Tests that never build or run:")
            for not_run in all_tests - to_be_run:
                print("- {}".format(not_run))

        return

    if options.subset:
        # Test instances are sorted depending on the context. For CI runs
        # the execution order is: "plat1-testA, plat1-testB, ...,
        # plat1-testZ, plat2-testA, ...". For hardware tests
        # (device_testing), were multiple physical platforms can run the tests
        # in parallel, it is more efficient to run in the order:
        # "plat1-testA, plat2-testA, ..., plat1-testB, plat2-testB, ..."
        if options.device_testing:
            suite.instances = OrderedDict(sorted(suite.instances.items(),
                                key=lambda x: x[0][x[0].find("/") + 1:]))
        else:
            suite.instances = OrderedDict(sorted(suite.instances.items()))

        # Do calculation based on what is actually going to be run and evaluated
        # at runtime, ignore the cases we already know going to be skipped.
        # This fixes an issue where some sets would get majority of skips and
        # basically run nothing beside filtering.
        to_run = {k : v for k,v in suite.instances.items() if v.status is None}

        subset, sets = options.subset.split("/")
        subset = int(subset)
        sets = int(sets)
        total = len(to_run)
        per_set = int(total / sets)
        num_extra_sets = total - (per_set * sets)

        # Try and be more fair for rounding error with integer division
        # so the last subset doesn't get overloaded, we add 1 extra to
        # subsets 1..num_extra_sets.
        if subset <= num_extra_sets:
            start = (subset - 1) * (per_set + 1)
            end = start + per_set + 1
        else:
            base = num_extra_sets * (per_set + 1)
            start = ((subset - num_extra_sets - 1) * per_set) + base
            end = start + per_set

        sliced_instances = islice(to_run.items(), start, end)
        skipped = {k : v for k,v in suite.instances.items() if v.status == 'skipped'}
        errors = {k : v for k,v in suite.instances.items() if v.status == 'error'}
        suite.instances = OrderedDict(sliced_instances)
        if subset == 1:
            # add all pre-filtered tests that are skipped or got error status
            # to the first set to allow for better distribution among all sets.
            suite.instances.update(skipped)
            suite.instances.update(errors)

    if options.save_tests:
        suite.csv_report(options.save_tests)
        return

    logger.info("%d test scenarios (%d configurations) selected, %d configurations discarded due to filters." %
                (len(suite.testcases), len(suite.instances), len(discards)))

    if options.device_testing and not options.build_only:
        print("\nDevice testing on:")
        hwm.dump(filtered=suite.selected_platforms)
        print("")

    if options.dry_run:
        duration = time.time() - start_time
        logger.info("Completed in %d seconds" % (duration))
        return

    retries = options.retry_failed + 1
    completed = 0

    BaseManager.register('LifoQueue', queue.LifoQueue)
    manager = BaseManager()
    manager.start()

    results = ExecutionCounter(total=len(suite.instances))
    pipeline = manager.LifoQueue()
    done_queue = manager.LifoQueue()

    suite.update_counting(results, initial=True)
    suite.start_time = start_time

    while True:
        completed += 1

        if completed > 1:
            logger.info("%d Iteration:" % (completed))
            time.sleep(options.retry_interval)  # waiting for the system to settle down
            results.done = results.total - results.failed
            results.failed = results.error

        results = suite.execute(pipeline, done_queue, results)

        while True:
            try:
                inst = done_queue.get_nowait()
            except queue.Empty:
                break
            else:
                inst.metrics.update(suite.instances[inst.name].metrics)
                inst.metrics["handler_time"] = inst.handler.duration if inst.handler else 0
                inst.metrics["unrecognized"] = []
                suite.instances[inst.name] = inst

        print("")

        retries = retries - 1
        # There are cases where failed == error (only build failures),
        # we do not try build failures.
        if retries == 0 or results.failed == results.error:
            break

    # figure out which report to use for size comparison
    if options.compare_report:
        report_to_use = options.compare_report
    elif options.last_metrics:
        report_to_use = previous_results_file
    else:
        report_to_use = suite.RELEASE_DATA

    suite.footprint_reports(report_to_use,
                            options.show_footprint,
                            options.all_deltas,
                            options.footprint_threshold,
                            options.last_metrics)

    suite.duration = time.time() - start_time
    suite.update_counting(results)

    suite.summary(results, options.disable_unrecognized_section_test)

    if options.coverage:
        if not options.gcov_tool:
            use_system_gcov = False

            for plat in options.coverage_platform:
                ts_plat = suite.get_platform(plat)
                if ts_plat and (ts_plat.type in {"native", "unit"}):
                    use_system_gcov = True

            if use_system_gcov or "ZEPHYR_SDK_INSTALL_DIR" not in os.environ:
                options.gcov_tool = "gcov"
            else:
                options.gcov_tool = os.path.join(os.environ["ZEPHYR_SDK_INSTALL_DIR"],
                                                 "x86_64-zephyr-elf/bin/x86_64-zephyr-elf-gcov")

        logger.info("Generating coverage files...")
        coverage_tool = CoverageTool.factory(options.coverage_tool)
        coverage_tool.gcov_tool = options.gcov_tool
        coverage_tool.base_dir = os.path.abspath(options.coverage_basedir)
        coverage_tool.add_ignore_file('generated')
        coverage_tool.add_ignore_directory('tests')
        coverage_tool.add_ignore_directory('samples')
        coverage_tool.generate(options.outdir)

    if options.device_testing and not options.build_only:
        print("\nHardware distribution summary:\n")
        table = []
        header = ['Board', 'ID', 'Counter']
        for d in hwm.duts:
            if d.connected and d.platform in suite.selected_platforms:
                row = [d.platform, d.id, d.counter]
                table.append(row)
        print(tabulate(table, headers=header, tablefmt="github"))

    suite.save_reports(options.report_name,
                       options.report_suffix,
                       options.report_dir,
                       options.no_update,
                       options.release,
                       options.only_failed,
                       options.platform_reports,
                       options.json_report
                       )

    # FIXME: remove later
    #logger.info(f"failed: {results.failed}, cases: {results.cases}, skipped configurations: {results.skipped_configs}, skipped_cases: {results.skipped_cases}, skipped(runtime): {results.skipped_runtime}, passed: {results.passed}, total: {results.total}, done: {results.done}")

    logger.info("Run completed")
    if results.failed or (suite.warnings and options.warnings_as_errors):
        sys.exit(1)


if __name__ == "__main__":
    try:
        main()
    finally:
        if (os.name != "nt") and os.isatty(1):
            # (OS is not Windows) and (stdout is interactive)
            os.system("stty sane")