Source code

Revision control

Copy as Markdown

Other Tools

# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.
import errno
import getpass
import io
import json
import logging
import os
import re
import subprocess
import sys
import time
from collections import Counter, OrderedDict, namedtuple
from itertools import dropwhile, islice, takewhile
from textwrap import TextWrapper
import six
from mach.site import CommandSiteManager
try:
import psutil
except Exception:
psutil = None
import mozfile
import mozpack.path as mozpath
from mach.mixin.logging import LoggingMixin
from mach.util import get_state_dir, get_virtualenv_base_dir
from mozsystemmonitor.resourcemonitor import SystemResourceMonitor
from mozterm.widgets import Footer
from ..backend import get_backend_class
from ..base import MozbuildObject
from ..compilation.warnings import WarningsCollector, WarningsDatabase
from ..dirutils import mkdir
from ..telemetry import get_cpu_brand
from ..testing import install_test_files
from ..util import FileAvoidWrite, resolve_target_to_make
from .clobber import Clobberer
FINDER_SLOW_MESSAGE = """
===================
PERFORMANCE WARNING
The OS X Finder application (file indexing used by Spotlight) used a lot of CPU
during the build - an average of %f%% (100%% is 1 core). This made your build
slower.
Consider adding ".noindex" to the end of your object directory name to have
Finder ignore it. Or, add an indexing exclusion through the Spotlight System
Preferences.
===================
""".strip()
INSTALL_TESTS_CLOBBER = "".join(
[
TextWrapper().fill(line) + "\n"
for line in """
The build system was unable to install tests because the CLOBBER file has \
been updated. This means if you edited any test files, your changes may not \
be picked up until a full/clobber build is performed.
The easiest and fastest way to perform a clobber build is to run:
$ mach clobber
$ mach build
If you did not modify any test files, it is safe to ignore this message \
and proceed with running tests. To do this run:
$ touch {clobber_file}
""".splitlines()
]
)
CLOBBER_REQUESTED_MESSAGE = """
===================
The CLOBBER file was updated prior to this build. A clobber build may be
required to succeed, but we weren't expecting it to.
Please consider filing a bug for this failure if you have reason to believe
this is a clobber bug and not due to local changes.
===================
""".strip()
BuildOutputResult = namedtuple(
"BuildOutputResult", ("warning", "state_changed", "message")
)
class TierStatus(object):
"""Represents the state and progress of tier traversal.
The build system is organized into linear phases called tiers. Each tier
executes in the order it was defined, 1 at a time.
"""
def __init__(self, resources):
"""Accepts a SystemResourceMonitor to record results against."""
self.tiers = OrderedDict()
self.tier_status = OrderedDict()
self.resources = resources
def set_tiers(self, tiers):
"""Record the set of known tiers."""
for tier in tiers:
self.tiers[tier] = dict(
begin_time=None,
finish_time=None,
duration=None,
)
self.tier_status[tier] = None
def begin_tier(self, tier):
"""Record that execution of a tier has begun."""
self.tier_status[tier] = "active"
t = self.tiers[tier]
t["begin_time"] = time.monotonic()
self.resources.begin_phase(tier)
def finish_tier(self, tier):
"""Record that execution of a tier has finished."""
self.tier_status[tier] = "finished"
t = self.tiers[tier]
t["finish_time"] = time.monotonic()
t["duration"] = self.resources.finish_phase(tier)
def record_cargo_timings(resource_monitor, timings_path):
cargo_start = 0
try:
with open(timings_path) as fh:
# Extrace the UNIT_DATA list from the cargo timing HTML file.
unit_data = dropwhile(lambda l: l.rstrip() != "const UNIT_DATA = [", fh)
unit_data = islice(unit_data, 1, None)
lines = takewhile(lambda l: l.rstrip() != "];", unit_data)
entries = json.loads("[" + "".join(lines) + "]")
# Normalize the entries so that any change in data format would
# trigger the exception handler that skips this (we don't want the
# build to fail in that case)
data = [
(
"{} v{}{}".format(
entry["name"], entry["version"], entry.get("target", "")
),
entry["start"] or 0,
entry["duration"] or 0,
)
for entry in entries
]
starts = [
start
for marker, start in resource_monitor._active_markers.items()
if marker.startswith("Rust:")
]
# The build system is not supposed to be running more than one cargo
# at the same time, which thankfully makes it easier to find the start
# of the one we got the timings for.
if len(starts) != 1:
return
cargo_start = starts[0]
except Exception:
return
if not cargo_start:
return
for name, start, duration in data:
resource_monitor.record_marker(
"RustCrate", cargo_start + start, cargo_start + start + duration, name
)
class BuildMonitor(MozbuildObject):
"""Monitors the output of the build."""
def init(self, warnings_path, terminal):
"""Create a new monitor.
warnings_path is a path of a warnings database to use.
"""
self._warnings_path = warnings_path
self.resources = SystemResourceMonitor(
poll_interval=0.1,
metadata={"CPUName": get_cpu_brand()},
)
self._resources_started = False
self.tiers = TierStatus(self.resources)
self.warnings_database = WarningsDatabase()
if os.path.exists(warnings_path):
try:
self.warnings_database.load_from_file(warnings_path)
except ValueError:
os.remove(warnings_path)
# Contains warnings unique to this invocation. Not populated with old
# warnings.
self.instance_warnings = WarningsDatabase()
self._terminal = terminal
def on_warning(warning):
# Skip `errors`
if warning["type"] == "error":
return
filename = warning["filename"]
if not os.path.exists(filename):
raise Exception("Could not find file containing warning: %s" % filename)
self.warnings_database.insert(warning)
# Make a copy so mutations don't impact other database.
self.instance_warnings.insert(warning.copy())
self._warnings_collector = WarningsCollector(on_warning, objdir=self.topobjdir)
self.build_objects = []
self.build_dirs = set()
def start(self):
"""Record the start of the build."""
self.start_time = time.monotonic()
self._finder_start_cpu = self._get_finder_cpu_usage()
def start_resource_recording(self):
# This should be merged into start() once bug 892342 lands.
self.resources.start()
self._resources_started = True
def on_line(self, line):
"""Consume a line of output from the build system.
This will parse the line for state and determine whether more action is
needed.
Returns a BuildOutputResult instance.
In this named tuple, warning will be an object describing a new parsed
warning. Otherwise it will be None.
state_changed indicates whether the build system changed state with
this line. If the build system changed state, the caller may want to
query this instance for the current state in order to update UI, etc.
message is either None, or the content of a message to be
displayed to the user.
"""
message = None
# If the previous line was colored (eg. for a compiler warning), our
# line will start with the ansi reset sequence. Strip it to ensure it
# does not interfere with our parsing of the line.
plain_line = self._terminal.strip(line) if self._terminal else line.strip()
if plain_line.startswith("BUILDSTATUS"):
args = plain_line.split()
_, _, disambiguator = args.pop(0).partition("@")
action = args.pop(0)
update_needed = True
if action == "TIERS":
self.tiers.set_tiers(args)
update_needed = False
elif action == "TIER_START":
tier = args[0]
self.tiers.begin_tier(tier)
elif action == "TIER_FINISH":
(tier,) = args
self.tiers.finish_tier(tier)
elif action == "OBJECT_FILE":
self.build_objects.append(args[0])
self.resources.begin_marker("Object", args[0], disambiguator)
update_needed = False
elif action.startswith("START_"):
self.resources.begin_marker(
action[len("START_") :], " ".join(args), disambiguator
)
update_needed = False
elif action.startswith("END_"):
self.resources.end_marker(
action[len("END_") :], " ".join(args), disambiguator
)
update_needed = False
elif action == "BUILD_VERBOSE":
build_dir = args[0]
if build_dir not in self.build_dirs:
self.build_dirs.add(build_dir)
message = build_dir
update_needed = False
else:
raise Exception("Unknown build status: %s" % action)
return BuildOutputResult(None, update_needed, message)
elif plain_line.startswith("Timing report saved to "):
cargo_timings = plain_line[len("Timing report saved to ") :]
record_cargo_timings(self.resources, cargo_timings)
return BuildOutputResult(None, False, None)
warning = None
message = line
try:
warning = self._warnings_collector.process_line(line)
except Exception:
pass
return BuildOutputResult(warning, False, message)
def stop_resource_recording(self):
if self._resources_started:
self.resources.stop()
self._resources_started = False
def finish(self):
"""Record the end of the build."""
self.stop_resource_recording()
self.end_time = time.monotonic()
self._finder_end_cpu = self._get_finder_cpu_usage()
self.elapsed = self.end_time - self.start_time
self.warnings_database.prune()
self.warnings_database.save_to_file(self._warnings_path)
def record_usage(self):
build_resources_profile_path = None
try:
# When running on automation, we store the resource usage data in
# the upload path, alongside, for convenience, a copy of the HTML
# viewer.
if "MOZ_AUTOMATION" in os.environ and "UPLOAD_PATH" in os.environ:
build_resources_profile_path = mozpath.join(
os.environ["UPLOAD_PATH"], "profile_build_resources.json"
)
else:
build_resources_profile_path = self._get_state_filename(
"profile_build_resources.json"
)
with io.open(
build_resources_profile_path, "w", encoding="utf-8", newline="\n"
) as fh:
to_write = six.ensure_text(
json.dumps(self.resources.as_profile(), separators=(",", ":"))
)
fh.write(to_write)
except Exception as e:
self.log(
logging.WARNING,
"build_resources_error",
{"msg": str(e)},
"Exception when writing resource usage file: {msg}",
)
try:
if build_resources_profile_path and os.path.exists(
build_resources_profile_path
):
os.remove(build_resources_profile_path)
except Exception:
# In case there's an exception for some reason, ignore it.
pass
def _get_finder_cpu_usage(self):
"""Obtain the CPU usage of the Finder app on OS X.
This is used to detect high CPU usage.
"""
if not sys.platform.startswith("darwin"):
return None
if not psutil:
return None
for proc in psutil.process_iter():
if proc.name != "Finder":
continue
if proc.username != getpass.getuser():
continue
# Try to isolate system finder as opposed to other "Finder"
# processes.
if not proc.exe.endswith("CoreServices/Finder.app/Contents/MacOS/Finder"):
continue
return proc.get_cpu_times()
return None
def have_high_finder_usage(self):
"""Determine whether there was high Finder CPU usage during the build.
Returns True if there was high Finder CPU usage, False if there wasn't,
or None if there is nothing to report.
"""
if not self._finder_start_cpu:
return None, None
# We only measure if the measured range is sufficiently long.
if self.elapsed < 15:
return None, None
if not self._finder_end_cpu:
return None, None
start = self._finder_start_cpu
end = self._finder_end_cpu
start_total = start.user + start.system
end_total = end.user + end.system
cpu_seconds = end_total - start_total
# If Finder used more than 25% of 1 core during the build, report an
# error.
finder_percent = cpu_seconds / self.elapsed * 100
return finder_percent > 25, finder_percent
def have_excessive_swapping(self):
"""Determine whether there was excessive swapping during the build.
Returns a tuple of (excessive, swap_in, swap_out). All values are None
if no swap information is available.
"""
if not self.have_resource_usage:
return None, None, None
swap_in = sum(m.swap.sin for m in self.resources.measurements)
swap_out = sum(m.swap.sout for m in self.resources.measurements)
# The threshold of 1024 MB has been arbitrarily chosen.
#
# Choosing a proper value that is ideal for everyone is hard. We will
# likely iterate on the logic until people are generally satisfied.
# If a value is too low, the eventual warning produced does not carry
# much meaning. If the threshold is too high, people may not see the
# warning and the warning will thus be ineffective.
excessive = swap_in > 512 * 1048576 or swap_out > 512 * 1048576
return excessive, swap_in, swap_out
@property
def have_resource_usage(self):
"""Whether resource usage is available."""
return self.resources.start_time is not None
def get_resource_usage(self):
"""Produce a data structure containing the low-level resource usage information.
This data structure can e.g. be serialized into JSON and saved for
subsequent analysis.
If no resource usage is available, None is returned.
"""
if not self.have_resource_usage:
return None
cpu_percent = self.resources.aggregate_cpu_percent(phase=None, per_cpu=False)
io = self.resources.aggregate_io(phase=None)
return dict(
cpu_percent=cpu_percent,
io=io,
)
def log_resource_usage(self, usage):
"""Summarize the resource usage of this build in a log message."""
if not usage:
return
params = dict(
duration=self.end_time - self.start_time,
cpu_percent=usage["cpu_percent"],
io_read_bytes=usage["io"].read_bytes,
io_write_bytes=usage["io"].write_bytes,
)
message = (
"Overall system resources - Wall time: {duration:.0f}s; "
"CPU: {cpu_percent:.0f}%; "
"Read bytes: {io_read_bytes}; Write bytes: {io_write_bytes}; "
)
if hasattr(usage["io"], "read_time") and hasattr(usage["io"], "write_time"):
params.update(
io_read_time=usage["io"].read_time,
io_write_time=usage["io"].write_time,
)
message += "Read time: {io_read_time}; Write time: {io_write_time}"
self.log(logging.WARNING, "resource_usage", params, message)
excessive, sin, sout = self.have_excessive_swapping()
if excessive is not None and (sin or sout):
sin /= 1048576
sout /= 1048576
self.log(
logging.WARNING,
"swap_activity",
{"sin": sin, "sout": sout},
"Swap in/out (MB): {sin}/{sout}",
)
def ccache_stats(self, ccache=None):
ccache_stats = None
if ccache is None:
ccache = mozfile.which("ccache")
if ccache:
# With CCache v3.7+ we can use --print-stats
has_machine_format = CCacheStats.check_version_3_7_or_newer(ccache)
try:
output = subprocess.check_output(
[ccache, "--print-stats" if has_machine_format else "-s"],
universal_newlines=True,
)
ccache_stats = CCacheStats(output, has_machine_format)
except ValueError as e:
self.log(logging.WARNING, "ccache", {"msg": str(e)}, "{msg}")
return ccache_stats
class TerminalLoggingHandler(logging.Handler):
"""Custom logging handler that works with terminal window dressing.
This class should probably live elsewhere, like the mach core. Consider
this a proving ground for its usefulness.
"""
def __init__(self):
logging.Handler.__init__(self)
self.fh = sys.stdout
self.footer = None
def flush(self):
self.acquire()
try:
self.fh.flush()
finally:
self.release()
def emit(self, record):
msg = self.format(record)
self.acquire()
try:
if self.footer:
self.footer.clear()
self.fh.write(msg)
self.fh.write("\n")
if self.footer:
self.footer.draw()
# If we don't flush, the footer may not get drawn.
self.fh.flush()
finally:
self.release()
class BuildProgressFooter(Footer):
"""Handles display of a build progress indicator in a terminal.
When mach builds inside a blessed-supported terminal, it will render
progress information collected from a BuildMonitor. This class converts the
state of BuildMonitor into terminal output.
"""
def __init__(self, terminal, monitor):
Footer.__init__(self, terminal)
self.tiers = six.viewitems(monitor.tiers.tier_status)
def draw(self):
"""Draws this footer in the terminal."""
if not self.tiers:
return
# The drawn terminal looks something like:
# TIER: static export libs tools
parts = [("bold", "TIER:")]
append = parts.append
for tier, status in self.tiers:
if status is None:
append(tier)
elif status == "finished":
append(("green", tier))
else:
append(("underline_yellow", tier))
self.write(parts)
class OutputManager(LoggingMixin):
"""Handles writing job output to a terminal or log."""
def __init__(self, log_manager, footer):
self.populate_logger()
self.footer = None
terminal = log_manager.terminal
# TODO convert terminal footer to config file setting.
if not terminal:
return
if os.environ.get("INSIDE_EMACS", None):
return
if os.environ.get("MACH_NO_TERMINAL_FOOTER", None):
footer = None
self.t = terminal
self.footer = footer
self._handler = TerminalLoggingHandler()
self._handler.setFormatter(log_manager.terminal_formatter)
self._handler.footer = self.footer
old = log_manager.replace_terminal_handler(self._handler)
self._handler.level = old.level
def __enter__(self):
return self
def __exit__(self, exc_type, exc_value, traceback):
if self.footer:
self.footer.clear()
# Prevents the footer from being redrawn if logging occurs.
self._handler.footer = None
def write_line(self, line):
if self.footer:
self.footer.clear()
print(line)
if self.footer:
self.footer.draw()
def refresh(self):
if not self.footer:
return
self.footer.clear()
self.footer.draw()
class BuildOutputManager(OutputManager):
"""Handles writing build output to a terminal, to logs, etc."""
def __init__(self, log_manager, monitor, footer):
self.monitor = monitor
OutputManager.__init__(self, log_manager, footer)
def __exit__(self, exc_type, exc_value, traceback):
OutputManager.__exit__(self, exc_type, exc_value, traceback)
# Ensure the resource monitor is stopped because leaving it running
# could result in the process hanging on exit because the resource
# collection child process hasn't been told to stop.
self.monitor.stop_resource_recording()
def on_line(self, line):
warning, state_changed, message = self.monitor.on_line(line)
if message:
self.log(logging.INFO, "build_output", {"line": message}, "{line}")
elif state_changed:
have_handler = hasattr(self, "_handler")
if have_handler:
self._handler.acquire()
try:
self.refresh()
finally:
if have_handler:
self._handler.release()
class StaticAnalysisFooter(Footer):
"""Handles display of a static analysis progress indicator in a terminal."""
def __init__(self, terminal, monitor):
Footer.__init__(self, terminal)
self.monitor = monitor
def draw(self):
"""Draws this footer in the terminal."""
monitor = self.monitor
total = monitor.num_files
processed = monitor.num_files_processed
percent = "(%.2f%%)" % (processed * 100.0 / total)
parts = [
("bright_black", "Processing"),
("yellow", str(processed)),
("bright_black", "of"),
("yellow", str(total)),
("bright_black", "files"),
("green", percent),
]
if monitor.current_file:
parts.append(("bold", monitor.current_file))
self.write(parts)
class StaticAnalysisOutputManager(OutputManager):
"""Handles writing static analysis output to a terminal or file."""
def __init__(self, log_manager, monitor, footer):
self.monitor = monitor
self.raw = ""
OutputManager.__init__(self, log_manager, footer)
def on_line(self, line):
warning, relevant = self.monitor.on_line(line)
if relevant:
self.raw += line + "\n"
if warning:
self.log(
logging.INFO,
"compiler_warning",
warning,
"Warning: {flag} in {filename}: {message}",
)
if relevant:
self.log(logging.INFO, "build_output", {"line": line}, "{line}")
else:
have_handler = hasattr(self, "_handler")
if have_handler:
self._handler.acquire()
try:
self.refresh()
finally:
if have_handler:
self._handler.release()
def write(self, path, output_format):
assert output_format in ("text", "json"), "Invalid output format {}".format(
output_format
)
path = mozpath.realpath(path)
if output_format == "json":
self.monitor._warnings_database.save_to_file(path)
else:
with io.open(path, "w", encoding="utf-8", newline="\n") as f:
f.write(self.raw)
self.log(
logging.INFO,
"write_output",
{"path": path, "format": output_format},
"Wrote {format} output in {path}",
)
class CCacheStats(object):
"""Holds statistics from ccache.
Instances can be subtracted from each other to obtain differences.
print() or str() the object to show a ``ccache -s`` like output
of the captured stats.
"""
STATS_KEYS = [
# (key, description)
# Refer to stats.c in ccache project for all the descriptions.
("stats_zeroed", ("stats zeroed", "stats zero time")),
("stats_updated", "stats updated"),
("cache_hit_direct", "cache hit (direct)"),
("cache_hit_preprocessed", "cache hit (preprocessed)"),
("cache_hit_rate", "cache hit rate"),
("cache_miss", "cache miss"),
("link", "called for link"),
("preprocessing", "called for preprocessing"),
("multiple", "multiple source files"),
("stdout", "compiler produced stdout"),
("no_output", "compiler produced no output"),
("empty_output", "compiler produced empty output"),
("failed", "compile failed"),
("error", "ccache internal error"),
("preprocessor_error", "preprocessor error"),
("cant_use_pch", "can't use precompiled header"),
("compiler_missing", "couldn't find the compiler"),
("cache_file_missing", "cache file missing"),
("bad_args", "bad compiler arguments"),
("unsupported_lang", "unsupported source language"),
("compiler_check_failed", "compiler check failed"),
("autoconf", "autoconf compile/link"),
("unsupported_code_directive", "unsupported code directive"),
("unsupported_compiler_option", "unsupported compiler option"),
("out_stdout", "output to stdout"),
("out_device", "output to a non-regular file"),
("no_input", "no input file"),
("bad_extra_file", "error hashing extra file"),
("num_cleanups", "cleanups performed"),
("cache_files", "files in cache"),
("cache_size", "cache size"),
("cache_max_size", "max cache size"),
]
SKIP_LINES = (
"cache directory",
"primary config",
"secondary config",
)
STATS_KEYS_3_7_PLUS = {
"stats_zeroed_timestamp": "stats_zeroed",
"stats_updated_timestamp": "stats_updated",
"direct_cache_hit": "cache_hit_direct",
"preprocessed_cache_hit": "cache_hit_preprocessed",
# "cache_hit_rate" is not provided
"cache_miss": "cache_miss",
"called_for_link": "link",
"called_for_preprocessing": "preprocessing",
"multiple_source_files": "multiple",
"compiler_produced_stdout": "stdout",
"compiler_produced_no_output": "no_output",
"compiler_produced_empty_output": "empty_output",
"compile_failed": "failed",
"internal_error": "error",
"preprocessor_error": "preprocessor_error",
"could_not_use_precompiled_header": "cant_use_pch",
"could_not_find_compiler": "compiler_missing",
"missing_cache_file": "cache_file_missing",
"bad_compiler_arguments": "bad_args",
"unsupported_source_language": "unsupported_lang",
"compiler_check_failed": "compiler_check_failed",
"autoconf_test": "autoconf",
"unsupported_code_directive": "unsupported_code_directive",
"unsupported_compiler_option": "unsupported_compiler_option",
"output_to_stdout": "out_stdout",
"output_to_a_non_file": "out_device",
"no_input_file": "no_input",
"error_hashing_extra_file": "bad_extra_file",
"cleanups_performed": "num_cleanups",
"files_in_cache": "cache_files",
"cache_size_kibibyte": "cache_size",
# "cache_max_size" is obsolete and not printed anymore
}
ABSOLUTE_KEYS = {"cache_files", "cache_size", "cache_max_size"}
FORMAT_KEYS = {"cache_size", "cache_max_size"}
GiB = 1024**3
MiB = 1024**2
KiB = 1024
def __init__(self, output=None, has_machine_format=False):
"""Construct an instance from the output of ccache -s."""
self._values = {}
if not output:
return
if has_machine_format:
self._parse_machine_format(output)
else:
self._parse_human_format(output)
def _parse_machine_format(self, output):
for line in output.splitlines():
line = line.strip()
key, _, value = line.partition("\t")
stat_key = self.STATS_KEYS_3_7_PLUS.get(key)
if stat_key:
value = int(value)
if key.endswith("_kibibyte"):
value *= 1024
self._values[stat_key] = value
(direct, preprocessed, miss) = self.hit_rates()
self._values["cache_hit_rate"] = (direct + preprocessed) * 100
def _parse_human_format(self, output):
for line in output.splitlines():
line = line.strip()
if line:
self._parse_line(line)
def _parse_line(self, line):
line = six.ensure_text(line)
for stat_key, stat_description in self.STATS_KEYS:
if line.startswith(stat_description):
raw_value = self._strip_prefix(line, stat_description)
self._values[stat_key] = self._parse_value(raw_value)
break
else:
if not line.startswith(self.SKIP_LINES):
raise ValueError("Failed to parse ccache stats output: %s" % line)
@staticmethod
def _strip_prefix(line, prefix):
if isinstance(prefix, tuple):
for p in prefix:
line = CCacheStats._strip_prefix(line, p)
return line
return line[len(prefix) :].strip() if line.startswith(prefix) else line
@staticmethod
def _parse_value(raw_value):
try:
# ccache calls strftime with '%c' (src/stats.c)
ts = time.strptime(raw_value, "%c")
return int(time.mktime(ts))
except ValueError:
if raw_value == "never":
return 0
pass
value = raw_value.split()
unit = ""
if len(value) == 1:
numeric = value[0]
elif len(value) == 2:
numeric, unit = value
else:
raise ValueError("Failed to parse ccache stats value: %s" % raw_value)
if "." in numeric:
numeric = float(numeric)
else:
numeric = int(numeric)
if unit in ("GB", "Gbytes"):
unit = CCacheStats.GiB
elif unit in ("MB", "Mbytes"):
unit = CCacheStats.MiB
elif unit in ("KB", "Kbytes"):
unit = CCacheStats.KiB
else:
unit = 1
return int(numeric * unit)
def hit_rate_message(self):
return (
"ccache (direct) hit rate: {:.1%}; (preprocessed) hit rate: {:.1%};"
" miss rate: {:.1%}".format(*self.hit_rates())
)
def hit_rates(self):
direct = self._values["cache_hit_direct"]
preprocessed = self._values["cache_hit_preprocessed"]
miss = self._values["cache_miss"]
total = float(direct + preprocessed + miss)
if total > 0:
direct /= total
preprocessed /= total
miss /= total
return (direct, preprocessed, miss)
def __sub__(self, other):
result = CCacheStats()
for k, prefix in self.STATS_KEYS:
if k not in self._values and k not in other._values:
continue
our_value = self._values.get(k, 0)
other_value = other._values.get(k, 0)
if k in self.ABSOLUTE_KEYS:
result._values[k] = our_value
else:
result._values[k] = our_value - other_value
return result
def __str__(self):
LEFT_ALIGN = 34
lines = []
for stat_key, stat_description in self.STATS_KEYS:
if stat_key not in self._values:
continue
value = self._values[stat_key]
if stat_key in self.FORMAT_KEYS:
value = "%15s" % self._format_value(value)
else:
value = "%8u" % value
if isinstance(stat_description, tuple):
stat_description = stat_description[0]
lines.append("%s%s" % (stat_description.ljust(LEFT_ALIGN), value))
return "\n".join(lines)
def __nonzero__(self):
relative_values = [
v for k, v in self._values.items() if k not in self.ABSOLUTE_KEYS
]
return all(v >= 0 for v in relative_values) and any(
v > 0 for v in relative_values
)
def __bool__(self):
return self.__nonzero__()
@staticmethod
def _format_value(v):
if v > CCacheStats.GiB:
return "%.1f Gbytes" % (float(v) / CCacheStats.GiB)
elif v > CCacheStats.MiB:
return "%.1f Mbytes" % (float(v) / CCacheStats.MiB)
else:
return "%.1f Kbytes" % (float(v) / CCacheStats.KiB)
@staticmethod
def check_version_3_7_or_newer(ccache):
output_version = subprocess.check_output(
[ccache, "--version"], universal_newlines=True
)
return CCacheStats._is_version_3_7_or_newer(output_version)
@staticmethod
def _is_version_3_7_or_newer(output):
if "ccache version" not in output:
return False
major = 0
minor = 0
for line in output.splitlines():
version = re.search(r"ccache version (\d+).(\d+).*", line)
if version:
major = int(version.group(1))
minor = int(version.group(2))
break
return ((major << 8) + minor) >= ((3 << 8) + 7)
class BuildDriver(MozbuildObject):
"""Provides a high-level API for build actions."""
def __init__(self, *args, **kwargs):
MozbuildObject.__init__(self, *args, virtualenv_name="build", **kwargs)
self.metrics = None
self.mach_context = None
def build(
self,
metrics,
what=None,
jobs=0,
job_size=0,
directory=None,
verbose=False,
keep_going=False,
mach_context=None,
append_env=None,
):
warnings_path = self._get_state_filename("warnings.json")
monitor = self._spawn(BuildMonitor)
monitor.init(warnings_path, self.log_manager.terminal)
status = self._build(
monitor,
metrics,
what,
jobs,
job_size,
directory,
verbose,
keep_going,
mach_context,
append_env,
)
record_usage = True
# On automation, only record usage for plain `mach build`
if "MOZ_AUTOMATION" in os.environ and what:
record_usage = False
if record_usage:
monitor.record_usage()
return status
def _build(
self,
monitor,
metrics,
what=None,
jobs=0,
job_size=0,
directory=None,
verbose=False,
keep_going=False,
mach_context=None,
append_env=None,
):
"""Invoke the build backend.
``what`` defines the thing to build. If not defined, the default
target is used.
"""
self.metrics = metrics
self.mach_context = mach_context
footer = BuildProgressFooter(self.log_manager.terminal, monitor)
# Disable indexing in objdir because it is not necessary and can slow
# down builds.
mkdir(self.topobjdir, not_indexed=True)
with BuildOutputManager(self.log_manager, monitor, footer) as output:
monitor.start()
if directory is not None and not what:
print("Can only use -C/--directory with an explicit target " "name.")
return 1
if directory is not None:
directory = mozpath.normsep(directory)
if directory.startswith("/"):
directory = directory[1:]
monitor.start_resource_recording()
if self._check_clobber(self.mozconfig, os.environ):
return 1
self.mach_context.command_attrs["clobber"] = False
self.metrics.mozbuild.clobber.set(False)
config = None
try:
config = self.config_environment
except Exception:
# If we don't already have a config environment this is either
# a fresh objdir or $OBJDIR/config.status has been removed for
# some reason, which indicates a clobber of sorts.
self.mach_context.command_attrs["clobber"] = True
self.metrics.mozbuild.clobber.set(True)
# Record whether a clobber was requested so we can print
# a special message later if the build fails.
clobber_requested = False
# Write out any changes to the current mozconfig in case
# they should invalidate configure.
self._write_mozconfig_json()
previous_backend = None
if config is not None:
previous_backend = config.substs.get("BUILD_BACKENDS", [None])[0]
config_rc = None
# Even if we have a config object, it may be out of date
# if something that influences its result has changed.
if config is None or self.build_out_of_date(
mozpath.join(self.topobjdir, "config.status"),
mozpath.join(self.topobjdir, "config_status_deps.in"),
):
if previous_backend and "Make" not in previous_backend:
clobber_requested = self._clobber_configure()
if config is None:
print(" Config object not found by mach.")
config_rc = self.configure(
metrics,
buildstatus_messages=True,
line_handler=output.on_line,
append_env=append_env,
)
if config_rc != 0:
return config_rc
config = self.reload_config_environment()
if config.substs.get("MOZ_USING_CCACHE"):
ccache = config.substs.get("CCACHE")
ccache_start = monitor.ccache_stats(ccache)
else:
ccache_start = None
# Collect glean metrics
substs = config.substs
mozbuild_metrics = metrics.mozbuild
mozbuild_metrics.compiler.set(substs.get("CC_TYPE", None))
def get_substs_flag(name):
return bool(substs.get(name, None))
host = substs.get("host")
monitor.resources.metadata["oscpu"] = host
target = substs.get("target")
if host != target:
monitor.resources.metadata["abi"] = target
product_name = substs.get("MOZ_BUILD_APP")
app_displayname = substs.get("MOZ_APP_DISPLAYNAME")
if app_displayname:
product_name = app_displayname
app_version = substs.get("MOZ_APP_VERSION")
if app_version:
product_name += " " + app_version
monitor.resources.metadata["product"] = product_name
mozbuild_metrics.artifact.set(get_substs_flag("MOZ_ARTIFACT_BUILDS"))
mozbuild_metrics.debug.set(get_substs_flag("MOZ_DEBUG"))
mozbuild_metrics.opt.set(get_substs_flag("MOZ_OPTIMIZE"))
mozbuild_metrics.ccache.set(get_substs_flag("CCACHE"))
using_sccache = get_substs_flag("MOZ_USING_SCCACHE")
mozbuild_metrics.sccache.set(using_sccache)
mozbuild_metrics.icecream.set(get_substs_flag("CXX_IS_ICECREAM"))
mozbuild_metrics.project.set(substs.get("MOZ_BUILD_APP", ""))
all_backends = config.substs.get("BUILD_BACKENDS", [None])
active_backend = all_backends[0]
status = None
if not config_rc and any(
[
self.backend_out_of_date(
mozpath.join(self.topobjdir, "backend.%sBackend" % backend)
)
for backend in all_backends
]
):
print("Build configuration changed. Regenerating backend.")
args = [
config.substs["PYTHON3"],
mozpath.join(self.topobjdir, "config.status"),
]
self.run_process(args, cwd=self.topobjdir, pass_thru=True)
if jobs == 0:
for param in self.mozconfig.get("make_extra") or []:
key, value = param.split("=", 1)
if key == "MOZ_PARALLEL_BUILD":
jobs = int(value)
if "Make" not in active_backend:
backend_cls = get_backend_class(active_backend)(config)
status = backend_cls.build(self, output, jobs, verbose, what)
if status and clobber_requested:
for line in CLOBBER_REQUESTED_MESSAGE.splitlines():
self.log(
logging.WARNING, "clobber", {"msg": line.rstrip()}, "{msg}"
)
if what and status is None:
# Collect target pairs.
target_pairs = []
for target in what:
path_arg = self._wrap_path_argument(target)
if directory is not None:
make_dir = mozpath.join(self.topobjdir, directory)
make_target = target
else:
make_dir, make_target = resolve_target_to_make(
self.topobjdir, path_arg.relpath()
)
if make_dir is None and make_target is None:
return 1
if config.is_artifact_build and target.startswith("installers-"):
print(
"Localized Builds are not supported with Artifact Builds enabled.\n"
"You should disable Artifact Builds (Use --disable-compile-environment "
"in your mozconfig instead) then re-build to proceed."
)
return 1
# See bug 886162 - we don't want to "accidentally" build
# the entire tree (if that's really the intent, it's
# unlikely they would have specified a directory.)
if not make_dir and not make_target:
print(
"The specified directory doesn't contain a "
"Makefile and the first parent with one is the "
"root of the tree. Please specify a directory "
"with a Makefile or run |mach build| if you "
"want to build the entire tree."
)
return 1
target_pairs.append((make_dir, make_target))
# Build target pairs.
for make_dir, make_target in target_pairs:
# We don't display build status messages during partial
# tree builds because they aren't reliable there. This
# could potentially be fixed if the build monitor were more
# intelligent about encountering undefined state.
no_build_status = "1" if make_dir is not None else ""
tgt_env = dict(append_env or {})
tgt_env["NO_BUILDSTATUS_MESSAGES"] = no_build_status
status = self._run_make(
directory=make_dir,
target=make_target,
line_handler=output.on_line,
log=False,
print_directory=False,
ensure_exit_code=False,
num_jobs=jobs,
job_size=job_size,
silent=not verbose,
append_env=tgt_env,
keep_going=keep_going,
)
if status != 0:
break
elif status is None:
# If the backend doesn't specify a build() method, then just
# call client.mk directly.
status = self._run_client_mk(
line_handler=output.on_line,
jobs=jobs,
job_size=job_size,
verbose=verbose,
keep_going=keep_going,
append_env=append_env,
)
self.log(
logging.WARNING,
"warning_summary",
{"count": len(monitor.warnings_database)},
"{count} compiler warnings present.",
)
# Try to run the active build backend's post-build step, if possible.
try:
active_backend = config.substs.get("BUILD_BACKENDS", [None])[0]
if active_backend:
backend_cls = get_backend_class(active_backend)(config)
new_status = backend_cls.post_build(
self, output, jobs, verbose, status
)
status = new_status
except Exception as ex:
self.log(
logging.DEBUG,
"post_build",
{"ex": str(ex)},
"Unable to run active build backend's post-build step; "
+ "failing the build due to exception: {ex}.",
)
if not status:
# If the underlying build provided a failing status, pass
# it through; otherwise, fail.
status = 1
monitor.finish()
if status == 0:
usage = monitor.get_resource_usage()
if usage:
self.mach_context.command_attrs["usage"] = usage
monitor.log_resource_usage(usage)
# Print the collected compiler warnings. This is redundant with
# inline output from the compiler itself. However, unlike inline
# output, this list is sorted and grouped by file, making it
# easier to triage output.
#
# Only do this if we had a successful build. If the build failed,
# there are more important things in the log to look for than
# whatever code we warned about.
if not status:
# Suppress warnings for 3rd party projects in local builds
# until we suppress them for real.
# TODO remove entries/feature once we stop generating warnings
# in these directories.
pathToThirdparty = mozpath.join(
self.topsrcdir, "tools", "rewriting", "ThirdPartyPaths.txt"
)
pathToGenerated = mozpath.join(
self.topsrcdir, "tools", "rewriting", "Generated.txt"
)
if os.path.exists(pathToThirdparty):
with io.open(
pathToThirdparty, encoding="utf-8", newline="\n"
) as f, io.open(pathToGenerated, encoding="utf-8", newline="\n") as g:
# Normalize the path (no trailing /)
LOCAL_SUPPRESS_DIRS = tuple(
[line.strip("\n/") for line in f]
+ [line.strip("\n/") for line in g]
)
else:
# For application based on gecko like thunderbird
LOCAL_SUPPRESS_DIRS = ()
suppressed_by_dir = Counter()
THIRD_PARTY_CODE = "third-party code"
suppressed = set(
w.replace("-Wno-error=", "-W")
for w in substs.get("WARNINGS_CFLAGS", [])
+ substs.get("WARNINGS_CXXFLAGS", [])
if w.startswith("-Wno-error=")
)
warnings = []
for warning in sorted(monitor.instance_warnings):
path = mozpath.normsep(warning["filename"])
if path.startswith(self.topsrcdir):
path = path[len(self.topsrcdir) + 1 :]
warning["normpath"] = path
if "MOZ_AUTOMATION" not in os.environ:
if path.startswith(LOCAL_SUPPRESS_DIRS):
suppressed_by_dir[THIRD_PARTY_CODE] += 1
continue
if warning["flag"] in suppressed:
suppressed_by_dir[mozpath.dirname(path)] += 1
continue
warnings.append(warning)
if THIRD_PARTY_CODE in suppressed_by_dir:
suppressed_third_party_code = [
(THIRD_PARTY_CODE, suppressed_by_dir.pop(THIRD_PARTY_CODE))
]
else:
suppressed_third_party_code = []
for d, count in suppressed_third_party_code + sorted(
suppressed_by_dir.items()
):
self.log(
logging.WARNING,
"suppressed_warning",
{"dir": d, "count": count},
"(suppressed {count} warnings in {dir})",
)
for warning in warnings:
if warning["column"] is not None:
self.log(
logging.WARNING,
"compiler_warning",
warning,
"warning: {normpath}:{line}:{column} [{flag}] " "{message}",
)
else:
self.log(
logging.WARNING,
"compiler_warning",
warning,
"warning: {normpath}:{line} [{flag}] {message}",
)
high_finder, finder_percent = monitor.have_high_finder_usage()
if high_finder:
print(FINDER_SLOW_MESSAGE % finder_percent)
if config.substs.get("MOZ_USING_CCACHE"):
ccache_end = monitor.ccache_stats(ccache)
else:
ccache_end = None
ccache_diff = None
if ccache_start and ccache_end:
ccache_diff = ccache_end - ccache_start
if ccache_diff:
self.log(
logging.INFO,
"ccache",
{"msg": ccache_diff.hit_rate_message()},
"{msg}",
)
notify_minimum_time = 300
try:
notify_minimum_time = int(os.environ.get("MACH_NOTIFY_MINTIME", "300"))
except ValueError:
# Just stick with the default
pass
if monitor.elapsed > notify_minimum_time:
# Display a notification when the build completes.
self.notify("Build complete" if not status else "Build failed")
if status:
if what and any(
[target for target in what if target not in ("faster", "binaries")]
):
print(
"Hey! Builds initiated with `mach build "
"$A_SPECIFIC_TARGET` may not always work, even if the "
"code being built is correct. Consider doing a bare "
"`mach build` instead."
)
return status
if monitor.have_resource_usage:
excessive, swap_in, swap_out = monitor.have_excessive_swapping()
# if excessive:
# print(EXCESSIVE_SWAP_MESSAGE)
print("To view a profile of the build, run |mach " "resource-usage|.")
long_build = monitor.elapsed > 1200
if long_build:
output.on_line(
"We know it took a while, but your build finally finished successfully!"
)
if not using_sccache:
output.on_line(
"If you are building Firefox often, SCCache can save you a lot "
"of time. You can learn more here: "
"configuring_build_options.html#sccache"
)
else:
output.on_line("Your build was successful!")
# Only for full builds because incremental builders likely don't
# need to be burdened with this.
if not what:
try:
# Fennec doesn't have useful output from just building. We should
# arguably make the build action useful for Fennec. Another day...
if self.substs["MOZ_BUILD_APP"] != "mobile/android":
print("To take your build for a test drive, run: |mach run|")
app = self.substs["MOZ_BUILD_APP"]
if app in ("browser", "mobile/android"):
print(
"For more information on what to do now, see "
)
except Exception:
# Ignore Exceptions in case we can't find config.status (such
# as when doing OSX Universal builds)
pass
return status
def configure(
self,
metrics,
options=None,
buildstatus_messages=False,
line_handler=None,
append_env=None,
):
# Disable indexing in objdir because it is not necessary and can slow
# down builds.
self.metrics = metrics
mkdir(self.topobjdir, not_indexed=True)
self._write_mozconfig_json()
def on_line(line):
self.log(logging.INFO, "build_output", {"line": line}, "{line}")
line_handler = line_handler or on_line
append_env = dict(append_env or {})
# Back when client.mk was used, `mk_add_options "export ..."` lines
# from the mozconfig would spill into the configure environment, so
# add that for backwards compatibility.
for line in self.mozconfig["make_extra"] or []:
if line.startswith("export "):
k, eq, v = line[len("export ") :].partition("=")
if eq == "=":
append_env[k] = v
build_site = CommandSiteManager.from_environment(
self.topsrcdir,
lambda: get_state_dir(specific_to_topsrcdir=True, topsrcdir=self.topsrcdir),
"build",
get_virtualenv_base_dir(self.topsrcdir),
)
build_site.ensure()
command = [build_site.python_path, mozpath.join(self.topsrcdir, "configure.py")]
if options:
command.extend(options)
if buildstatus_messages:
append_env["MOZ_CONFIGURE_BUILDSTATUS"] = "1"
line_handler("BUILDSTATUS TIERS configure")
line_handler("BUILDSTATUS TIER_START configure")
env = os.environ.copy()
env.update(append_env)
with subprocess.Popen(
command,
cwd=self.topobjdir,
env=env,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
universal_newlines=True,
) as process:
for line in process.stdout:
line_handler(line.rstrip())
status = process.wait()
if buildstatus_messages:
line_handler("BUILDSTATUS TIER_FINISH configure")
if status:
print('*** Fix above errors and then restart with "./mach build"')
else:
print("Configure complete!")
print("Be sure to run |mach build| to pick up any changes")
return status
def install_tests(self):
"""Install test files."""
if self.is_clobber_needed():
print(
INSTALL_TESTS_CLOBBER.format(
clobber_file=mozpath.join(self.topobjdir, "CLOBBER")
)
)
sys.exit(1)
install_test_files(mozpath.normpath(self.topsrcdir), self.topobjdir, "_tests")
def _clobber_configure(self):
# This is an optimistic treatment of the CLOBBER file for when we have
# some trust in the build system: an update to the CLOBBER file is
# interpreted to mean that configure will fail during an incremental
# build, which is handled by removing intermediate configure artifacts
# and subsections of the objdir related to python and testing before
# proceeding.
clobberer = Clobberer(self.topsrcdir, self.topobjdir)
clobber_output = io.StringIO()
res = clobberer.maybe_do_clobber(os.getcwd(), False, clobber_output)
required, performed, message = res
assert not performed
if not required:
return False
def remove_objdir_path(path):
path = mozpath.join(self.topobjdir, path)
self.log(