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 multiprocessing
import sys
import time
import warnings
from collections import OrderedDict, namedtuple
from contextlib import contextmanager
class PsutilStub(object):
def __init__(self):
self.sswap = namedtuple(
"sswap", ["total", "used", "free", "percent", "sin", "sout"]
)
self.sdiskio = namedtuple(
"sdiskio",
[
"read_count",
"write_count",
"read_bytes",
"write_bytes",
"read_time",
"write_time",
],
)
self.pcputimes = namedtuple("pcputimes", ["user", "system"])
self.svmem = namedtuple(
"svmem",
[
"total",
"available",
"percent",
"used",
"free",
"active",
"inactive",
"buffers",
"cached",
],
)
def cpu_count(self, logical=True):
return 0
def cpu_percent(self, a, b):
return [0]
def cpu_times(self, percpu):
if percpu:
return [self.pcputimes(0, 0)]
else:
return self.pcputimes(0, 0)
def disk_io_counters(self):
return self.sdiskio(0, 0, 0, 0, 0, 0)
def swap_memory(self):
return self.sswap(0, 0, 0, 0, 0, 0)
def virtual_memory(self):
return self.svmem(0, 0, 0, 0, 0, 0, 0, 0, 0)
# psutil will raise NotImplementedError if the platform is not supported.
try:
import psutil
have_psutil = True
except Exception:
try:
# The PsutilStub should get us time intervals, at least
psutil = PsutilStub()
except Exception:
psutil = None
have_psutil = False
def get_disk_io_counters():
try:
io_counters = psutil.disk_io_counters()
if io_counters is None:
return PsutilStub().disk_io_counters()
except RuntimeError:
io_counters = PsutilStub().disk_io_counters()
return io_counters
def _poll(pipe, poll_interval=0.1):
"""Wrap multiprocessing.Pipe.poll to hide POLLERR and POLLIN
exceptions.
multiprocessing.Pipe is not actually a pipe on at least Linux.
That has an effect on the expected outcome of reading from it when
the other end of the pipe dies, leading to possibly hanging on revc()
below.
"""
try:
return pipe.poll(poll_interval)
except Exception:
# Poll might throw an exception even though there's still
# data to read. That happens when the underlying system call
# returns both POLLERR and POLLIN, but python doesn't tell us
# about it. So assume there is something to read, and we'll
# get an exception when trying to read the data.
return True
def _collect(pipe, poll_interval):
"""Collects system metrics.
This is the main function for the background process. It collects
data then forwards it on a pipe until told to stop.
"""
data = []
try:
# Establish initial values.
io_last = get_disk_io_counters()
swap_last = psutil.swap_memory()
psutil.cpu_percent(None, True)
cpu_last = psutil.cpu_times(True)
last_time = time.monotonic()
sin_index = swap_last._fields.index("sin")
sout_index = swap_last._fields.index("sout")
sleep_interval = poll_interval
while not _poll(pipe, poll_interval=sleep_interval):
io = get_disk_io_counters()
virt_mem = psutil.virtual_memory()
swap_mem = psutil.swap_memory()
cpu_percent = psutil.cpu_percent(None, True)
cpu_times = psutil.cpu_times(True)
# Take the timestamp as soon as possible after getting cpu_times
# to reduce the likelihood of our process being interrupted between
# the two instructions. Having a delayed timestamp would cause the
# next sample to report more than 100% CPU time.
measured_end_time = time.monotonic()
# TODO Does this wrap? At 32 bits? At 64 bits?
# TODO Consider patching "delta" API to upstream.
io_diff = [v - io_last[i] for i, v in enumerate(io)]
io_last = io
cpu_diff = []
for core, values in enumerate(cpu_times):
cpu_diff.append([v - cpu_last[core][i] for i, v in enumerate(values)])
cpu_last = cpu_times
swap_entry = list(swap_mem)
swap_entry[sin_index] = swap_mem.sin - swap_last.sin
swap_entry[sout_index] = swap_mem.sout - swap_last.sout
swap_last = swap_mem
data.append(
(
last_time,
measured_end_time,
io_diff,
cpu_diff,
cpu_percent,
list(virt_mem),
swap_entry,
)
)
collection_overhead = time.monotonic() - last_time - sleep_interval
last_time = measured_end_time
sleep_interval = max(poll_interval / 2, poll_interval - collection_overhead)
except Exception as e:
warnings.warn("_collect failed: %s" % e)
finally:
for entry in data:
pipe.send(entry)
pipe.send(("done", None, None, None, None, None, None))
pipe.close()
sys.exit(0)
SystemResourceUsage = namedtuple(
"SystemResourceUsage",
["start", "end", "cpu_times", "cpu_percent", "io", "virt", "swap"],
)
class SystemResourceMonitor(object):
"""Measures system resources.
Each instance measures system resources from the time it is started
until it is finished. It does this on a separate process so it doesn't
impact execution of the main Python process.
Each instance is a one-shot instance. It cannot be used to record multiple
durations.
Aside from basic data gathering, the class supports basic analysis
capabilities. You can query for data between ranges. You can also tell it
when certain events occur and later grab data relevant to those events or
plot those events on a timeline.
The resource monitor works by periodically polling the state of the
system. By default, it polls every second. This can be adjusted depending
on the required granularity of the data and considerations for probe
overhead. It tries to probe at the interval specified. However, variations
should be expected. Fast and well-behaving systems should experience
variations in the 1ms range. Larger variations may exist if the system is
under heavy load or depending on how accurate socket polling is on your
system.
In its current implementation, data is not available until collection has
stopped. This may change in future iterations.
Usage
=====
monitor = SystemResourceMonitor()
monitor.start()
# Record that a single event in time just occurred.
foo.do_stuff()
monitor.record_event('foo_did_stuff')
# Record that we're about to perform a possibly long-running event.
with monitor.phase('long_job'):
foo.do_long_running_job()
# Stop recording. Currently we need to stop before data is available.
monitor.stop()
# Obtain the raw data for the entire probed range.
print('CPU Usage:')
for core in monitor.aggregate_cpu():
print(core)
# We can also request data corresponding to a specific phase.
for data in monitor.phase_usage('long_job'):
print(data.cpu_percent)
"""
# The interprocess communication is complicated enough to warrant
# explanation. To work around the Python GIL, we launch a separate
# background process whose only job is to collect metrics. If we performed
# collection in the main process, the polling interval would be
# inconsistent if a long-running function were on the stack. Since the
# child process is independent of the instantiating process, data
# collection should be evenly spaced.
#
# As the child process collects data, it buffers it locally. When
# collection stops, it flushes all that data to a pipe to be read by
# the parent process.
instance = None
def __init__(self, poll_interval=1.0, metadata={}):
"""Instantiate a system resource monitor instance.
The instance is configured with a poll interval. This is the interval
between samples, in float seconds.
"""
self.start_time = None
self.end_time = None
self.events = []
self.markers = []
self.phases = OrderedDict()
self._active_phases = {}
self._active_markers = {}
self._running = False
self._stopped = False
self._process = None
if psutil is None:
return
# This try..except should not be needed! However, some tools (like
# |mach build|) attempt to load psutil before properly creating a
# virtualenv by building psutil. As a result, python/psutil may be in
# sys.path and its .py files may pick up the psutil C extension from
# the system install. If the versions don't match, we typically see
# failures invoking one of these functions.
try:
cpu_percent = psutil.cpu_percent(0.0, True)
cpu_times = psutil.cpu_times(False)
io = get_disk_io_counters()
virt = psutil.virtual_memory()
swap = psutil.swap_memory()
except Exception as e:
warnings.warn("psutil failed to run: %s" % e)
return
self._cpu_cores = len(cpu_percent)
self._cpu_times_type = type(cpu_times)
self._cpu_times_len = len(cpu_times)
self._io_type = type(io)
self._io_len = len(io)
self._virt_type = type(virt)
self._virt_len = len(virt)
self._swap_type = type(swap)
self._swap_len = len(swap)
self.start_timestamp = time.time()
self.start_time = time.monotonic()
self._pipe, child_pipe = multiprocessing.Pipe(True)
self._process = multiprocessing.Process(
target=_collect, args=(child_pipe, poll_interval)
)
self.poll_interval = poll_interval
self.metadata = metadata
def __del__(self):
if self._running:
self._pipe.send(("terminate",))
self._process.join()
def convert_to_monotonic_time(self, timestamp):
return timestamp - self.start_timestamp + self.start_time
# Methods to control monitoring.
def start(self):
"""Start measuring system-wide CPU resource utilization.
You should only call this once per instance.
"""
if not self._process:
return
self._process.start()
self._running = True
self.start_time = time.monotonic()
SystemResourceMonitor.instance = self
def stop(self):
"""Stop measuring system-wide CPU resource utilization.
You should call this if and only if you have called start(). You should
always pair a stop() with a start().
Currently, data is not available until you call stop().
"""
if not self._process:
self._stopped = True
return
self.stop_time = time.monotonic()
assert not self._stopped
try:
self._pipe.send(("terminate",))
except Exception:
pass
self._stopped = True
self.measurements = []
# The child process will send each data sample over the pipe
# as a separate data structure. When it has finished sending
# samples, it sends a special "done" message to indicate it
# is finished.
while _poll(self._pipe, poll_interval=0.1):
try:
(
start_time,
end_time,
io_diff,
cpu_diff,
cpu_percent,
virt_mem,
swap_mem,
) = self._pipe.recv()
except Exception as e:
warnings.warn("failed to receive data: %s" % e)
# Assume we can't recover
break
# There should be nothing after the "done" message so
# terminate.
if start_time == "done":
break
try:
io = self._io_type(*io_diff)
virt = self._virt_type(*virt_mem)
swap = self._swap_type(*swap_mem)
cpu_times = [self._cpu_times_type(*v) for v in cpu_diff]
self.measurements.append(
SystemResourceUsage(
start_time, end_time, cpu_times, cpu_percent, io, virt, swap
)
)
except Exception:
# We also can't recover, but output the data that caused the exception
warnings.warn(
"failed to read the received data: %s"
% str(
(
start_time,
end_time,
io_diff,
cpu_diff,
cpu_percent,
virt_mem,
swap_mem,
)
)
)
break
# We establish a timeout so we don't hang forever if the child
# process has crashed.
if self._running:
self._process.join(10)
if self._process.is_alive():
self._process.terminate()
self._process.join(10)
self._running = False
SystemResourceUsage.instance = None
self.end_time = time.monotonic()
# Methods to record events alongside the monitored data.
@staticmethod
def record_event(name):
"""Record an event as occuring now.
Events are actions that occur at a specific point in time. If you are
looking for an action that has a duration, see the phase API below.
"""
if SystemResourceMonitor.instance:
SystemResourceMonitor.instance.events.append((time.monotonic(), name))
@staticmethod
def record_marker(name, start, end, text):
"""Record a marker with a duration and an optional text
Markers are typically used to record when a single command happened.
For actions with a longer duration that justifies tracking resource use
see the phase API below.
"""
if SystemResourceMonitor.instance:
SystemResourceMonitor.instance.markers.append((name, start, end, text))
@staticmethod
def begin_marker(name, text, disambiguator=None, timestamp=None):
if SystemResourceMonitor.instance:
id = name + ":" + text
if disambiguator:
id += ":" + disambiguator
SystemResourceMonitor.instance._active_markers[id] = (
SystemResourceMonitor.instance.convert_to_monotonic_time(timestamp)
if timestamp
else time.monotonic()
)
@staticmethod
def end_marker(name, text, disambiguator=None, timestamp=None):
if not SystemResourceMonitor.instance:
return
end = time.monotonic()
if timestamp:
end = SystemResourceMonitor.instance.convert_to_monotonic_time(timestamp)
id = name + ":" + text
if disambiguator:
id += ":" + disambiguator
if not id in SystemResourceMonitor.instance._active_markers:
return
start = SystemResourceMonitor.instance._active_markers.pop(id)
SystemResourceMonitor.instance.record_marker(name, start, end, text)
@contextmanager
def phase(self, name):
"""Context manager for recording an active phase."""
self.begin_phase(name)
yield
self.finish_phase(name)
def begin_phase(self, name):
"""Record the start of a phase.
Phases are actions that have a duration. Multiple phases can be active
simultaneously. Phases can be closed in any order.
Keep in mind that if phases occur in parallel, it will become difficult
to isolate resource utilization specific to individual phases.
"""
assert name not in self._active_phases
self._active_phases[name] = time.monotonic()
def finish_phase(self, name):
"""Record the end of a phase."""
assert name in self._active_phases
phase = (self._active_phases[name], time.monotonic())
self.phases[name] = phase
del self._active_phases[name]
return phase[1] - phase[0]
# Methods to query data.
def range_usage(self, start=None, end=None):
"""Obtain the usage data falling within the given time range.
This is a generator of SystemResourceUsage.
If no time range bounds are given, all data is returned.
"""
if not self._stopped or self.start_time is None:
return
if start is None:
start = self.start_time
if end is None:
end = self.end_time
for entry in self.measurements:
if entry.start < start:
continue
if entry.end > end:
break
yield entry
def phase_usage(self, phase):
"""Obtain usage data for a specific phase.
This is a generator of SystemResourceUsage.
"""
time_start, time_end = self.phases[phase]
return self.range_usage(time_start, time_end)
def between_events_usage(self, start_event, end_event):
"""Obtain usage data between two point events.
This is a generator of SystemResourceUsage.
"""
start_time = None
end_time = None
for t, name in self.events:
if name == start_event:
start_time = t
elif name == end_event:
end_time = t
if start_time is None:
raise Exception("Could not find start event: %s" % start_event)
if end_time is None:
raise Exception("Could not find end event: %s" % end_event)
return self.range_usage(start_time, end_time)
def aggregate_cpu_percent(self, start=None, end=None, phase=None, per_cpu=True):
"""Obtain the aggregate CPU percent usage for a range.
Returns a list of floats representing average CPU usage percentage per
core if per_cpu is True (the default). If per_cpu is False, return a
single percentage value.
By default this will return data for the entire instrumented interval.
If phase is defined, data for a named phase will be returned. If start
and end are defined, these times will be fed into range_usage().
"""
cpu = [[] for i in range(0, self._cpu_cores)]
if phase:
data = self.phase_usage(phase)
else:
data = self.range_usage(start, end)
for usage in data:
for i, v in enumerate(usage.cpu_percent):
cpu[i].append(v)
samples = len(cpu[0])
if not samples:
return 0
if per_cpu:
# pylint --py3k W1619
return [sum(x) / samples for x in cpu]
cores = [sum(x) for x in cpu]
# pylint --py3k W1619
return sum(cores) / len(cpu) / samples
def aggregate_cpu_times(self, start=None, end=None, phase=None, per_cpu=True):
"""Obtain the aggregate CPU times for a range.
If per_cpu is True (the default), this returns a list of named tuples.
Each tuple is as if it were returned by psutil.cpu_times(). If per_cpu
is False, this returns a single named tuple of the aforementioned type.
"""
empty = [0 for i in range(0, self._cpu_times_len)]
cpu = [list(empty) for i in range(0, self._cpu_cores)]
if phase:
data = self.phase_usage(phase)
else:
data = self.range_usage(start, end)
for usage in data:
for i, core_values in enumerate(usage.cpu_times):
for j, v in enumerate(core_values):
cpu[i][j] += v
if per_cpu:
return [self._cpu_times_type(*v) for v in cpu]
sums = list(empty)
for core in cpu:
for i, v in enumerate(core):
sums[i] += v
return self._cpu_times_type(*sums)
def aggregate_io(self, start=None, end=None, phase=None):
"""Obtain aggregate I/O counters for a range.
Returns an iostat named tuple from psutil.
"""
io = [0 for i in range(self._io_len)]
if phase:
data = self.phase_usage(phase)
else:
data = self.range_usage(start, end)
for usage in data:
for i, v in enumerate(usage.io):
io[i] += v
return self._io_type(*io)
def min_memory_available(self, start=None, end=None, phase=None):
"""Return the minimum observed available memory number from a range.
Returns long bytes of memory available.
See psutil for notes on how this is calculated.
"""
if phase:
data = self.phase_usage(phase)
else:
data = self.range_usage(start, end)
values = []
for usage in data:
values.append(usage.virt.available)
return min(values)
def max_memory_percent(self, start=None, end=None, phase=None):
"""Returns the maximum percentage of system memory used.
Returns a float percentage. 1.00 would mean all system memory was in
use at one point.
"""
if phase:
data = self.phase_usage(phase)
else:
data = self.range_usage(start, end)
values = []
for usage in data:
values.append(usage.virt.percent)
return max(values)
def as_dict(self):
"""Convert the recorded data to a dict, suitable for serialization.
The returned dict has the following keys:
version - Integer version number being rendered. Currently 2.
cpu_times_fields - A list of the names of the CPU times fields.
io_fields - A list of the names of the I/O fields.
virt_fields - A list of the names of the virtual memory fields.
swap_fields - A list of the names of the swap memory fields.
samples - A list of dicts containing low-level measurements.
events - A list of lists representing point events. The inner list
has 2 elements, the float wall time of the event and the string
event name.
phases - A list of dicts describing phases. Each phase looks a lot
like an entry from samples (see below). Some phases may not have
data recorded against them, so some keys may be None.
overall - A dict representing overall resource usage. This resembles
a sample entry.
system - Contains additional information about the system including
number of processors and amount of memory.
Each entry in the sample list is a dict with the following keys:
start - Float wall time this measurement began on.
end - Float wall time this measurement ended on.
io - List of numerics for I/O values.
virt - List of numerics for virtual memory values.
swap - List of numerics for swap memory values.
cpu_percent - List of floats representing CPU percent on each core.
cpu_times - List of lists. Main list is each core. Inner lists are
lists of floats representing CPU times on that core.
cpu_percent_mean - Float of mean CPU percent across all cores.
cpu_times_sum - List of floats representing the sum of CPU times
across all cores.
cpu_times_total - Float representing the sum of all CPU times across
all cores. This is useful for calculating the percent in each CPU
time.
"""
o = dict(
version=2,
cpu_times_fields=list(self._cpu_times_type._fields),
io_fields=list(self._io_type._fields),
virt_fields=list(self._virt_type._fields),
swap_fields=list(self._swap_type._fields),
samples=[],
phases=[],
system={},
)
def populate_derived(e):
if e["cpu_percent_cores"]:
# pylint --py3k W1619
e["cpu_percent_mean"] = sum(e["cpu_percent_cores"]) / len(
e["cpu_percent_cores"]
)
else:
e["cpu_percent_mean"] = None
if e["cpu_times"]:
e["cpu_times_sum"] = [0.0] * self._cpu_times_len
for i in range(0, self._cpu_times_len):
e["cpu_times_sum"][i] = sum(core[i] for core in e["cpu_times"])
e["cpu_times_total"] = sum(e["cpu_times_sum"])
def phase_entry(name, start, end):
e = dict(
name=name,
start=start,
end=end,
duration=end - start,
cpu_percent_cores=self.aggregate_cpu_percent(phase=name),
cpu_times=[list(c) for c in self.aggregate_cpu_times(phase=name)],
io=list(self.aggregate_io(phase=name)),
)
populate_derived(e)
return e
for m in self.measurements:
e = dict(
start=m.start,
end=m.end,
io=list(m.io),
virt=list(m.virt),
swap=list(m.swap),
cpu_percent_cores=list(m.cpu_percent),
cpu_times=list(list(cpu) for cpu in m.cpu_times),
)
populate_derived(e)
o["samples"].append(e)
if o["samples"]:
o["start"] = o["samples"][0]["start"]
o["end"] = o["samples"][-1]["end"]
o["duration"] = o["end"] - o["start"]
o["overall"] = phase_entry(None, o["start"], o["end"])
else:
o["start"] = None
o["end"] = None
o["duration"] = None
o["overall"] = None
o["events"] = [list(ev) for ev in self.events]
for phase, v in self.phases.items():
o["phases"].append(phase_entry(phase, v[0], v[1]))
if have_psutil:
o["system"].update(
dict(
cpu_logical_count=psutil.cpu_count(logical=True),
cpu_physical_count=psutil.cpu_count(logical=False),
swap_total=psutil.swap_memory()[0],
vmem_total=psutil.virtual_memory()[0],
)
)
return o
def as_profile(self):
profile_time = time.monotonic()
start_time = self.start_time
profile = {
"meta": {
"processType": 0,
"product": "mach",
"stackwalk": 0,
"version": 27,
"preprocessedProfileVersion": 47,
"symbolicationNotSupported": True,
"interval": self.poll_interval * 1000,
"startTime": self.start_timestamp * 1000,
"profilingStartTime": 0,
"logicalCPUs": psutil.cpu_count(logical=True),
"physicalCPUs": psutil.cpu_count(logical=False),
"mainMemory": psutil.virtual_memory()[0],
"markerSchema": [
{
"name": "Phase",
"tooltipLabel": "{marker.data.phase}",
"tableLabel": "{marker.name} — {marker.data.phase} — CPU time: {marker.data.cpuTime} ({marker.data.cpuPercent})",
"chartLabel": "{marker.data.phase}",
"display": [
"marker-chart",
"marker-table",
"timeline-overview",
],
"data": [
{
"key": "cpuTime",
"label": "CPU Time",
"format": "duration",
},
{
"key": "cpuPercent",
"label": "CPU Percent",
"format": "string",
},
],
},
{
"name": "Text",
"tooltipLabel": "{marker.name}",
"tableLabel": "{marker.name} — {marker.data.text}",
"chartLabel": "{marker.data.text}",
"display": ["marker-chart", "marker-table"],
"data": [
{
"key": "text",
"label": "Description",
"format": "string",
"searchable": True,
}
],
},
{
"name": "Mem",
"tooltipLabel": "{marker.name}",
"display": [],
"data": [
{"key": "used", "label": "Memory Used", "format": "bytes"},
{
"key": "cached",
"label": "Memory cached",
"format": "bytes",
},
{
"key": "buffers",
"label": "Memory buffers",
"format": "bytes",
},
],
"graphs": [
{"key": "used", "color": "orange", "type": "line-filled"}
],
},
{
"name": "IO",
"tooltipLabel": "{marker.name}",
"display": [],
"data": [
{
"key": "write_bytes",
"label": "Written",
"format": "bytes",
},
{
"key": "write_count",
"label": "Write count",
"format": "integer",
},
{"key": "read_bytes", "label": "Read", "format": "bytes"},
{
"key": "read_count",
"label": "Read count",
"format": "integer",
},
],
"graphs": [
{"key": "read_bytes", "color": "green", "type": "bar"},
{"key": "write_bytes", "color": "red", "type": "bar"},
],
},
],
"usesOnlyOneStackType": True,
},
"libs": [],
"threads": [
{
"processType": "default",
"processName": "mach",
"processStartupTime": 0,
"processShutdownTime": None,
"registerTime": 0,
"unregisterTime": None,
"pausedRanges": [],
"showMarkersInTimeline": True,
"name": "",
"isMainThread": False,
"pid": "0",
"tid": 0,
"samples": {
"weightType": "samples",
"weight": None,
"stack": [],
"time": [],
"length": 0,
},
"stringArray": ["(root)"],
"markers": {
"data": [],
"name": [],
"startTime": [],
"endTime": [],
"phase": [],
"category": [],
"length": 0,
},
"stackTable": {
"frame": [0],
"prefix": [None],
"category": [0],
"subcategory": [0],
"length": 1,
},
"frameTable": {
"address": [-1],
"inlineDepth": [0],
"category": [None],
"subcategory": [0],
"func": [0],
"nativeSymbol": [None],
"innerWindowID": [0],
"implementation": [None],
"line": [None],
"column": [None],
"length": 1,
},
"funcTable": {
"isJS": [False],
"relevantForJS": [False],
"name": [0],
"resource": [-1],
"fileName": [None],
"lineNumber": [None],
"columnNumber": [None],
"length": 1,
},
"resourceTable": {
"lib": [],
"name": [],
"host": [],
"type": [],
"length": 0,
},
"nativeSymbols": {
"libIndex": [],
"address": [],
"name": [],
"functionSize": [],
"length": 0,
},
}
],
"counters": [],
}
firstThread = profile["threads"][0]
markers = firstThread["markers"]
for key in self.metadata:
profile["meta"][key] = self.metadata[key]
def get_string_index(string):
stringArray = firstThread["stringArray"]
try:
return stringArray.index(string)
except ValueError:
stringArray.append(string)
return len(stringArray) - 1
def add_marker(name_index, start, end, data, precision=None):
# The precision argument allows setting how many digits after the
# decimal point are desired.
# For resource use samples where we sample with a timer, an integer
# number of ms is good enough.
# For short duration markers, the profiler front-end may show up to
# 3 digits after the decimal point (ie. µs precision).
markers["startTime"].append(round((start - start_time) * 1000, precision))
if end is None:
markers["endTime"].append(None)
# 0 = Instant marker
markers["phase"].append(0)
else:
markers["endTime"].append(round((end - start_time) * 1000, precision))
# 1 = marker with start and end times, 2 = start but no end.
markers["phase"].append(1)
markers["category"].append(0)
markers["name"].append(name_index)
markers["data"].append(data)
markers["length"] = markers["length"] + 1
def format_percent(value):
return str(round(value, 1)) + "%"
samples = firstThread["samples"]
samples["stack"].append(0)
samples["time"].append(0)
cpu_string_index = get_string_index("CPU Use")
memory_string_index = get_string_index("Memory")
io_string_index = get_string_index("IO")
valid_cpu_fields = set()
for m in self.measurements:
# Ignore samples that are much too short.
if m.end - m.start < self.poll_interval / 10:
continue
# Sample times
samples["stack"].append(0)
samples["time"].append(round((m.end - start_time) * 1000))
# CPU
markerData = {
"type": "CPU",
"cpuPercent": format_percent(
sum(list(m.cpu_percent)) / len(m.cpu_percent)
),
}
# due to inconsistencies in the sampling rate, sometimes the
# cpu_times add up to more than 100%, causing annoying
# spikes in the CPU use charts. Avoid them by dividing the
# values by the total if it is above 1.
total = 0
for field in ["nice", "user", "system", "iowait", "softirq", "idle"]:
if hasattr(m.cpu_times[0], field):
total += sum(getattr(core, field) for core in m.cpu_times) / (
m.end - m.start
)
divisor = total if total > 1 else 1
total = 0
for field in ["nice", "user", "system", "iowait", "softirq"]:
if hasattr(m.cpu_times[0], field):
total += (
sum(getattr(core, field) for core in m.cpu_times)
/ (m.end - m.start)
/ divisor
)
if total > 0:
valid_cpu_fields.add(field)
markerData[field] = round(total, 3)
for field in ["nice", "user", "system", "iowait", "idle"]:
if hasattr(m.cpu_times[0], field):
markerData[field + "_pct"] = format_percent(
100
* sum(getattr(core, field) for core in m.cpu_times)
/ (m.end - m.start)
/ len(m.cpu_times)
)
add_marker(cpu_string_index, m.start, m.end, markerData)
# Memory
markerData = {"type": "Mem", "used": m.virt.used}
if hasattr(m.virt, "cached"):
markerData["cached"] = m.virt.cached
if hasattr(m.virt, "buffers"):
markerData["buffers"] = m.virt.buffers
add_marker(memory_string_index, m.start, m.end, markerData)
# IO
markerData = {
"type": "IO",
"read_count": m.io.read_count,
"read_bytes": m.io.read_bytes,
"write_count": m.io.write_count,
"write_bytes": m.io.write_bytes,
}
add_marker(io_string_index, m.start, m.end, markerData)
samples["length"] = len(samples["stack"])
# The marker schema for CPU markers should only contain graph
# definitions for fields we actually have, or the profiler front-end
# will detect missing data and skip drawing the track entirely.
cpuSchema = {
"name": "CPU",
"tooltipLabel": "{marker.name}",
"display": [],
"data": [{"key": "cpuPercent", "label": "CPU Percent", "format": "string"}],
"graphs": [],
}
cpuData = cpuSchema["data"]
for field, label in {
"user": "User %",
"iowait": "IO Wait %",
"system": "System %",
"nice": "Nice %",
"idle": "Idle %",
}.items():
if field in valid_cpu_fields or field == "idle":
cpuData.append(
{"key": field + "_pct", "label": label, "format": "string"}
)
cpuGraphs = cpuSchema["graphs"]
for field, color in {
"softirq": "orange",
"iowait": "red",
"system": "grey",
"user": "yellow",
"nice": "blue",
}.items():
if field in valid_cpu_fields:
cpuGraphs.append({"key": field, "color": color, "type": "bar"})
profile["meta"]["markerSchema"].insert(0, cpuSchema)
# Create markers for phases
phase_string_index = get_string_index("Phase")
for phase, v in self.phases.items():
markerData = {"type": "Phase", "phase": phase}
cpu_percent_cores = self.aggregate_cpu_percent(phase=phase)
if cpu_percent_cores:
markerData["cpuPercent"] = format_percent(
sum(cpu_percent_cores) / len(cpu_percent_cores)
)
cpu_times = [list(c) for c in self.aggregate_cpu_times(phase=phase)]
cpu_times_sum = [0.0] * self._cpu_times_len
for i in range(0, self._cpu_times_len):
cpu_times_sum[i] = sum(core[i] for core in cpu_times)
total_cpu_time_ms = sum(cpu_times_sum) * 1000
if total_cpu_time_ms > 0:
markerData["cpuTime"] = total_cpu_time_ms
add_marker(phase_string_index, v[0], v[1], markerData, 3)
# Add generic markers
for name, start, end, text in self.markers:
markerData = {"type": "Text"}
if text:
markerData["text"] = text
add_marker(get_string_index(name), start, end, markerData, 3)
if self.events:
event_string_index = get_string_index("Event")
for event_time, text in self.events:
if text:
add_marker(
event_string_index,
event_time,
None,
{"type": "Text", "text": text},
3,
)
# We may have spent some time generating this profile, and there might
# also have been some time elapsed between stopping the resource
# monitor, and the profile being created. These are hidden costs that
# we should account for as best as possible, and the best we can do
# is to make the profile contain information about this cost somehow.
# We extend the profile end time up to now rather than self.end_time,
# and add a phase covering that period of time.
now = time.monotonic()
profile["meta"]["profilingEndTime"] = round(
(now - self.start_time) * 1000 + 0.0005, 3
)
markerData = {
"type": "Phase",
"phase": "teardown",
}
add_marker(phase_string_index, self.stop_time, now, markerData, 3)
teardown_string_index = get_string_index("resourcemonitor")
markerData = {
"type": "Text",
"text": "stop",
}
add_marker(teardown_string_index, self.stop_time, self.end_time, markerData, 3)
markerData = {
"type": "Text",
"text": "as_profile",
}
add_marker(teardown_string_index, profile_time, now, markerData, 3)
# Unfortunately, whatever the caller does with the profile (e.g. json)
# or after that (hopefully, exit) is not going to be counted, but we
# assume it's fast enough.
return profile