blob: fd727c7f0922000e9645fd8153bd722c68fce2fa [file] [log] [blame]
# Copyright (c) 2011 The Chromium Authors. All rights reserved.
# Use of this source code is governed by a BSD-style license that can be
# found in the LICENSE file.
"""Runs an exe through Valgrind and puts the intermediate files in a
import datetime
import glob
import logging
import optparse
import os
import re
import shutil
import stat
import subprocess
import sys
import tempfile
import common
import drmemory_analyze
import memcheck_analyze
import tsan_analyze
class BaseTool(object):
"""Abstract class for running Valgrind-, PIN-based and other dynamic
error detector tools.
Always subclass this and implement ToolCommand with framework- and
tool-specific stuff.
def __init__(self):
self.temp_dir = tempfile.mkdtemp(prefix="vg_logs_") # Generated every time
self.log_dir = self.temp_dir # overridable by --keep_logs
self.option_parser_hooks = []
# TODO(glider): we may not need some of the env vars on some of the
# platforms.
self._env = {
"G_SLICE" : "always-malloc",
def ToolName(self):
raise NotImplementedError, "This method should be implemented " \
"in the tool-specific subclass"
def Analyze(self, check_sanity=False):
raise NotImplementedError, "This method should be implemented " \
"in the tool-specific subclass"
def RegisterOptionParserHook(self, hook):
# Frameworks and tools can add their own flags to the parser.
def CreateOptionParser(self):
# Defines Chromium-specific flags.
self._parser = optparse.OptionParser("usage: %prog [options] <program to "
self._parser.add_option("-t", "--timeout",
dest="timeout", metavar="TIMEOUT", default=10000,
help="timeout in seconds for the run (default 10000)")
self._parser.add_option("", "--source_dir",
help="path to top of source tree for this build"
"(used to normalize source paths in baseline)")
self._parser.add_option("", "--gtest_filter", default="",
help="which test case to run")
self._parser.add_option("", "--gtest_repeat",
help="how many times to run each test")
self._parser.add_option("", "--gtest_print_time", action="store_true",
help="show how long each test takes")
self._parser.add_option("", "--ignore_exit_code", action="store_true",
help="ignore exit code of the test "
"(e.g. test failures)")
self._parser.add_option("", "--keep_logs", action="store_true",
help="store memory tool logs in the <tool>.logs "
"directory instead of /tmp.\nThis can be "
"useful for tool developers/maintainers.\n"
"Please note that the <tool>.logs directory "
"will be clobbered on tool startup.")
# To add framework- or tool-specific flags, please add a hook using
# RegisterOptionParserHook in the corresponding subclass.
# See ValgrindTool and ThreadSanitizerBase for examples.
for hook in self.option_parser_hooks:
hook(self, self._parser)
def ParseArgv(self, args):
# self._tool_flags will store those tool flags which we don't parse
# manually in this script.
self._tool_flags = []
known_args = []
""" We assume that the first argument not starting with "-" is a program
name and all the following flags should be passed to the program.
TODO(timurrrr): customize optparse instead
while len(args) > 0 and args[0][:1] == "-":
arg = args[0]
if (arg == "--"):
if self._parser.has_option(arg.split("=")[0]):
known_args += [arg]
self._tool_flags += [arg]
args = args[1:]
if len(args) > 0:
known_args += args
self._options, self._args = self._parser.parse_args(known_args)
self._timeout = int(self._options.timeout)
self._source_dir = self._options.source_dir
if self._options.keep_logs:
self.log_dir = "%s.logs" % self.ToolName()
if os.path.exists(self.log_dir):
self._ignore_exit_code = self._options.ignore_exit_code
if self._options.gtest_filter != "":
self._args.append("--gtest_filter=%s" % self._options.gtest_filter)
if self._options.gtest_repeat:
self._args.append("--gtest_repeat=%s" % self._options.gtest_repeat)
if self._options.gtest_print_time:
return True
def Setup(self, args):
return self.ParseArgv(args)
def ToolCommand(self):
raise NotImplementedError, "This method should be implemented " \
"in the tool-specific subclass"
def Cleanup(self):
# You may override it in the tool-specific subclass
def Execute(self):
""" Execute the app to be tested after successful instrumentation.
Full execution command-line provided by subclassers via proc.""""starting execution...")
proc = self.ToolCommand()
for var in self._env:
common.PutEnvAndLog(var, self._env[var])
return common.RunSubprocess(proc, self._timeout)
def RunTestsAndAnalyze(self, check_sanity):
exec_retcode = self.Execute()
analyze_retcode = self.Analyze(check_sanity)
if analyze_retcode:
logging.error("Analyze failed.")"Search the log for '[ERROR]' to see the error reports.")
return analyze_retcode
if exec_retcode:
if self._ignore_exit_code:"Test execution failed, but the exit code is ignored.")
logging.error("Test execution failed.")
return exec_retcode
else:"Test execution completed successfully.")
if not analyze_retcode:"Analysis completed successfully.")
return 0
def Main(self, args, check_sanity):
"""Call this to run through the whole process: Setup, Execute, Analyze"""
start =
retcode = -1
if self.Setup(args):
retcode = self.RunTestsAndAnalyze(check_sanity)
shutil.rmtree(self.temp_dir, ignore_errors=True)
logging.error("Setup failed")
end =
seconds = (end - start).seconds
hours = seconds / 3600
seconds = seconds % 3600
minutes = seconds / 60
seconds = seconds % 60"elapsed time: %02d:%02d:%02d" % (hours, minutes, seconds))
return retcode
def Run(self, args, module):
# TODO(timurrrr): this is a temporary workaround for
if self.ToolName() == "tsan" and common.IsMac():
check_sanity = module in MODULES_TO_SANITY_CHECK
return self.Main(args, check_sanity)
class ValgrindTool(BaseTool):
"""Abstract class for running Valgrind tools.
Always subclass this and implement ToolSpecificFlags() and
ExtendOptionParser() for tool-specific stuff.
def __init__(self):
super(ValgrindTool, self).__init__()
def UseXML(self):
# Override if tool prefers nonxml output
return True
def SelfContained(self):
# Returns true iff the tool is distibuted as a self-contained
# .sh script (e.g. ThreadSanitizer)
return False
def ExtendOptionParser(self, parser):
parser.add_option("", "--suppressions", default=[],
help="path to a valgrind suppression file")
parser.add_option("", "--indirect", action="store_true",
help="set BROWSER_WRAPPER rather than "
"running valgrind directly")
parser.add_option("", "--indirect_webkit_layout", action="store_true",
help="set --wrapper rather than running valgrind "
parser.add_option("", "--trace_children", action="store_true",
help="also trace child processes")
parser.add_option("", "--num-callers",
dest="num_callers", default=30,
help="number of callers to show in stack traces")
parser.add_option("", "--generate_dsym", action="store_true",
help="Generate .dSYM file on Mac if needed. Slow!")
def Setup(self, args):
if not BaseTool.Setup(self, args):
return False
if common.IsMac():
return True
def PrepareForTestMac(self):
"""Runs dsymutil if needed.
Valgrind for Mac OS X requires that debugging information be in a .dSYM
bundle generated by dsymutil. It is not currently able to chase DWARF
data into .o files like gdb does, so executables without .dSYM bundles or
with the Chromium-specific "fake_dsym" bundles generated by
build/mac/strip_save_dsym won't give source file and line number
information in valgrind.
This function will run dsymutil if the .dSYM bundle is missing or if
it looks like a fake_dsym. A non-fake dsym that already exists is assumed
to be up-to-date.
test_command = self._args[0]
dsym_bundle = self._args[0] + '.dSYM'
dsym_file = os.path.join(dsym_bundle, 'Contents', 'Resources', 'DWARF',
dsym_info_plist = os.path.join(dsym_bundle, 'Contents', 'Info.plist')
needs_dsymutil = True
saved_test_command = None
if os.path.exists(dsym_file) and os.path.exists(dsym_info_plist):
# Look for the special fake_dsym tag in dsym_info_plist.
dsym_info_plist_contents = open(dsym_info_plist).read()
if not'^\s*<key>fake_dsym</key>$', dsym_info_plist_contents,
# fake_dsym is not set, this is a real .dSYM bundle produced by
# dsymutil. dsymutil does not need to be run again.
needs_dsymutil = False
# fake_dsym is set. dsym_file is a copy of the original test_command
# before it was stripped. Copy it back to test_command so that
# dsymutil has unstripped input to work with. Move the stripped
# test_command out of the way, it will be restored when this is
# done.
saved_test_command = test_command + '.stripped'
os.rename(test_command, saved_test_command)
shutil.copyfile(dsym_file, test_command)
shutil.copymode(saved_test_command, test_command)
if needs_dsymutil:
if self._options.generate_dsym:
# Remove the .dSYM bundle if it exists.
shutil.rmtree(dsym_bundle, True)
dsymutil_command = ['dsymutil', test_command]
# dsymutil is crazy slow. Ideally we'd have a timeout here,
# but common.RunSubprocess' timeout is only checked
# after each line of output; dsymutil is silent
# until the end, and is then killed, which is silly.
if saved_test_command:
os.rename(saved_test_command, test_command)
else:"No real .dSYM for test_command. Line numbers will "
"not be shown. Either tell xcode to generate .dSYM "
"file, or use --generate_dsym option to this tool.")
def ToolCommand(self):
"""Get the valgrind command to run."""
# Note that self._args begins with the exe to be run.
tool_name = self.ToolName()
# Construct the valgrind command.
if self.SelfContained():
proc = ["" % tool_name]
proc = ["valgrind", "--tool=%s" % tool_name]
proc += ["--num-callers=%i" % int(self._options.num_callers)]
if self._options.trace_children:
proc += ["--trace-children=yes"]
proc += self.ToolSpecificFlags()
proc += self._tool_flags
suppression_count = 0
for suppression_file in self._options.suppressions:
if os.path.exists(suppression_file):
suppression_count += 1
proc += ["--suppressions=%s" % suppression_file]
if not suppression_count:
logfilename = self.log_dir + ("/%s." % tool_name) + "%p"
if self.UseXML():
proc += ["--xml=yes", "--xml-file=" + logfilename]
proc += ["--log-file=" + logfilename]
# The Valgrind command is constructed.
# Handle --indirect_webkit_layout separately.
if self._options.indirect_webkit_layout:
# Need to create the wrapper before modifying |proc|.
wrapper = self.CreateBrowserWrapper(proc)
proc = self._args
return proc
if self._options.indirect:
wrapper = self.CreateBrowserWrapper(proc)
os.putenv("BROWSER_WRAPPER", wrapper)'export BROWSER_WRAPPER=' + wrapper)
proc = []
proc += self._args
return proc
def ToolSpecificFlags(self):
raise NotImplementedError, "This method should be implemented " \
"in the tool-specific subclass"
def CreateBrowserWrapper(self, proc):
"""The program being run invokes Python or something else that can't stand
to be valgrinded, and also invokes the Chrome browser. In this case, use a
magic wrapper to only valgrind the Chrome browser. Build the wrapper here.
Returns the path to the wrapper. It's up to the caller to use the wrapper
command = " ".join(proc)
# Add the PID of the browser wrapper to the logfile names so we can
# separate log files for different UI tests at the analyze stage.
command = command.replace("%p", "$$.%p")
(fd, indirect_fname) = tempfile.mkstemp(dir=self.log_dir,
f = os.fdopen(fd, "w")
'echo "Started Valgrind wrapper for this test, PID=$$"\n')
# Try to get the test case name by looking at the program arguments.
# i.e. Chromium ui_tests and friends pass --test-name arg.
f.write('DIR=`dirname $0`\n'
'for arg in $@; do\n'
' # TODO(timurrrr): this doesn\'t handle "--test-name Test.Name"\n'
' if [[ "$arg" =~ --test-name=(.*) ]]; then\n'
' fi\n'
f.write('if [ "$FOUND_TESTNAME" = "1" ]; then\n'
' %s "$@"\n'
'else\n' % command)
# Webkit layout_tests print out the test URL as the first line of stdout.
f.write(' %s "$@" | tee $DIR/test.$$.stdout\n'
' EXITCODE=$PIPESTATUS\n' # $? holds the tee's exit code
' head -n 1 $DIR/test.$$.stdout |\n'
' grep URL |\n'
' sed "s/^.*third_party\/WebKit\/LayoutTests\///" '
' exit $EXITCODE\n'
'fi\n' % command)
os.chmod(indirect_fname, stat.S_IRUSR|stat.S_IXUSR)
return indirect_fname
def CreateAnalyzer(self):
raise NotImplementedError, "This method should be implemented " \
"in the tool-specific subclass"
def GetAnalyzeResults(self, check_sanity=False):
# Glob all the files in the "testing.tmp" directory
filenames = glob.glob(self.log_dir + "/" + self.ToolName() + ".*")
# If we have browser wrapper, the logfiles are named as
# "toolname.wrapper_PID.valgrind_PID".
# Let's extract the list of wrapper_PIDs and name it ppids
ppids = set([int(f.split(".")[-2]) \
for f in filenames if"\.[0-9]+\.[0-9]+$", f)])
analyzer = self.CreateAnalyzer()
if len(ppids) == 0:
# Fast path - no browser wrapper was set.
return analyzer.Report(filenames, None, check_sanity)
ret = 0
for ppid in ppids:
testcase_name = None
f = open(self.log_dir + ("/" % ppid))
testcase_name =
except IOError:
print "====================================================="
print " Below is the report for valgrind wrapper PID=%d." % ppid
if testcase_name:
print " It was used while running the `%s` test." % testcase_name
print " You can find the corresponding test"
print " by searching the above log for 'PID=%d'" % ppid
ppid_filenames = [f for f in filenames \
if"\.%d\.[0-9]+$" % ppid, f)]
# check_sanity won't work with browser wrappers
assert check_sanity == False
ret |= analyzer.Report(ppid_filenames, testcase_name)
print "====================================================="
if ret != 0:
print ""
print "The Valgrind reports are grouped by test names."
print "Each test has its PID printed in the log when the test was run"
print "and at the beginning of its Valgrind report."
print "Hint: you can search for the reports by Ctrl+F -> `=#`"
return ret
# TODO(timurrrr): Split into a separate file.
class Memcheck(ValgrindTool):
Dynamic memory error detector for Linux & Mac
def __init__(self):
super(Memcheck, self).__init__()
def ToolName(self):
return "memcheck"
def ExtendOptionParser(self, parser):
parser.add_option("--leak-check", "--leak_check", type="string",
default="yes", # --leak-check=yes is equivalent of =full
help="perform leak checking at the end of the run")
parser.add_option("", "--show_all_leaks", action="store_true",
help="also show less blatant leaks")
parser.add_option("", "--track_origins", action="store_true",
help="Show whence uninitialized bytes came. 30% slower.")
def ToolSpecificFlags(self):
ret = ["--gen-suppressions=all", "--demangle=no"]
ret += ["--leak-check=%s" % self._options.leak_check]
if self._options.show_all_leaks:
ret += ["--show-reachable=yes"]
ret += ["--show-possibly-lost=no"]
if self._options.track_origins:
ret += ["--track-origins=yes"]
# TODO(glider): this is a temporary workaround for
# Let's see whether it helps.
if common.IsMac():
ret += ["--smc-check=all"]
return ret
def CreateAnalyzer(self):
use_gdb = common.IsMac()
return memcheck_analyze.MemcheckAnalyzer(self._source_dir,
def Analyze(self, check_sanity=False):
ret = self.GetAnalyzeResults(check_sanity)
if ret != 0:"Please see"
"using-valgrind for the info on Memcheck/Valgrind")
return ret
class PinTool(BaseTool):
"""Abstract class for running PIN tools.
Always subclass this and implement ToolSpecificFlags() and
ExtendOptionParser() for tool-specific stuff.
def PrepareForTest(self):
def ToolSpecificFlags(self):
raise NotImplementedError, "This method should be implemented " \
"in the tool-specific subclass"
def ToolCommand(self):
"""Get the PIN command to run."""
# Construct the PIN command.
pin_cmd = os.getenv("PIN_COMMAND")
if not pin_cmd:
raise RuntimeError, "Please set PIN_COMMAND environment variable " \
"with the path to pin.exe"
proc = pin_cmd.split(" ")
proc += self.ToolSpecificFlags()
# The PIN command is constructed.
# PIN requires -- to separate PIN flags from the executable name.
# self._args begins with the exe to be run.
proc += ["--"]
proc += self._args
return proc
class ThreadSanitizerBase(object):
Dynamic data race detector for Linux, Mac and Windows.
Since TSan works on both Valgrind (Linux, Mac) and PIN (Windows), we need
to have multiple inheritance
INFO_MESSAGE="Please see" \
"using-valgrind/threadsanitizer for the info on " \
def __init__(self):
super(ThreadSanitizerBase, self).__init__()
def ToolName(self):
return "tsan"
def UseXML(self):
return False
def SelfContained(self):
return True
def ExtendOptionParser(self, parser):
parser.add_option("", "--hybrid", default="no",
help="Finds more data races, may give false positive "
"reports unless the code is annotated")
parser.add_option("", "--announce-threads", default="yes",
help="Show the the stack traces of thread creation")
parser.add_option("", "--free-is-write", default="no",
help="Treat free()/operator delete as memory write. "
"This helps finding more data races, but (currently) "
"this may give false positive reports on std::string "
"internals, see"
def EvalBoolFlag(self, flag_value):
if (flag_value in ["1", "true", "yes"]):
return True
elif (flag_value in ["0", "false", "no"]):
return False
raise RuntimeError, "Can't parse flag value (%s)" % flag_value
def ToolSpecificFlags(self):
ret = []
ignore_files = ["ignores.txt"]
for platform_suffix in common.PlatformNames():
ignore_files.append("ignores_%s.txt" % platform_suffix)
for ignore_file in ignore_files:
fullname = os.path.join(self._source_dir,
"tools", "valgrind", "tsan", ignore_file)
if os.path.exists(fullname):
ret += ["--ignore=%s" % fullname]
# This should shorten filepaths for local builds.
ret += ["--file-prefix-to-cut=%s/" % self._source_dir]
# This should shorten filepaths on bots.
ret += ["--file-prefix-to-cut=build/src/"]
# This should shorten filepaths for functions intercepted in TSan.
ret += ["--file-prefix-to-cut=scripts/tsan/tsan/"]
ret += ["--gen-suppressions=true"]
if self.EvalBoolFlag(self._options.hybrid):
ret += ["--hybrid=yes"] # "no" is the default value for TSAN
if self.EvalBoolFlag(self._options.announce_threads):
ret += ["--announce-threads"]
if self.EvalBoolFlag(self._options.free_is_write):
ret += ["--free-is-write=yes"]
ret += ["--free-is-write=no"]
# --show-pc flag is needed for parsing the error logs on Darwin.
if platform_suffix == 'mac':
ret += ["--show-pc=yes"]
ret += ["--show-pid=no"]
boring_callers = common.BoringCallers(mangled=False, use_re_wildcards=False)
# TODO(timurrrr): In fact, we want "starting from .." instead of "below .."
for bc in boring_callers:
ret += ["--cut_stack_below=%s" % bc]
return ret
class ThreadSanitizerPosix(ThreadSanitizerBase, ValgrindTool):
def ToolSpecificFlags(self):
proc = ThreadSanitizerBase.ToolSpecificFlags(self)
# The -v flag is needed for printing the list of used suppressions and
# obtaining addresses for loaded shared libraries on Mac.
proc += ["-v"]
return proc
def CreateAnalyzer(self):
use_gdb = common.IsMac()
return tsan_analyze.TsanAnalyzer(self._source_dir, use_gdb)
def Analyze(self, check_sanity=False):
ret = self.GetAnalyzeResults(check_sanity)
if ret != 0:
return ret
class ThreadSanitizerWindows(ThreadSanitizerBase, PinTool):
def __init__(self):
super(ThreadSanitizerWindows, self).__init__()
def ExtendOptionParser(self, parser):
parser.add_option("", "--suppressions", default=[],
help="path to TSan suppression file")
def ToolSpecificFlags(self):
proc = ThreadSanitizerBase.ToolSpecificFlags(self)
# On PIN, ThreadSanitizer has its own suppression mechanism
# and --log-file flag which work exactly on Valgrind.
suppression_count = 0
for suppression_file in self._options.suppressions:
if os.path.exists(suppression_file):
suppression_count += 1
proc += ["--suppressions=%s" % suppression_file]
if not suppression_count:
logfilename = self.log_dir + "/tsan.%p"
proc += ["--log-file=" + logfilename]
# TODO(timurrrr): Add flags for Valgrind trace children analog when we
# start running complex tests (e.g. UI) under TSan/Win.
return proc
def Analyze(self, check_sanity=False):
filenames = glob.glob(self.log_dir + "/tsan.*")
analyzer = tsan_analyze.TsanAnalyzer(self._source_dir)
ret = analyzer.Report(filenames, None, check_sanity)
if ret != 0:
return ret
class DrMemory(BaseTool):
Dynamic memory error detector for Windows.
It is not very mature at the moment, some things might not work properly.
def __init__(self, handle_uninits_and_leaks):
super(DrMemory, self).__init__()
self.handle_uninits_and_leaks = handle_uninits_and_leaks
def ToolName(self):
return "DrMemory"
def ExtendOptionParser(self, parser):
parser.add_option("", "--suppressions", default=[],
help="path to a drmemory suppression file")
parser.add_option("", "--follow_python", action="store_true",
default=False, dest="follow_python",
help="Monitor python child processes. If off, neither "
"python children nor any children of python children "
"will be monitored.")
parser.add_option("", "--indirect", action="store_true",
help="set BROWSER_WRAPPER rather than "
"running Dr. Memory directly on the harness")
parser.add_option("", "--use_debug", action="store_true",
default=False, dest="use_debug",
help="Run Dr. Memory debug build")
parser.add_option("", "--trace_children", action="store_true",
help="TODO: default value differs from Valgrind")
def ToolCommand(self):
"""Get the tool command to run."""
tool_name = self.ToolName()
# WINHEAP is what Dr. Memory supports as there are issues w/ both
# jemalloc ( and
# tcmalloc (
add_env = {
for k,v in add_env.iteritems():"export %s=%s", k, v)
os.putenv(k, v)
drmem_cmd = os.getenv("DRMEMORY_COMMAND")
if not drmem_cmd:
raise RuntimeError, "Please set DRMEMORY_COMMAND environment variable " \
"with the path to drmemory.exe"
proc = drmem_cmd.split(" ")
# By default, don't run python (this will exclude python's children as well)
# to reduce runtime. We're not really interested in spending time finding
# bugs in the python implementation.
# With file-based config we must update the file every time, and
# it will affect simultaneous drmem uses by this user. While file-based
# config has many advantages, here we may want this-instance-only
# (
drconfig_cmd = [ proc[0].replace("drmemory.exe", "drconfig.exe") ]
drconfig_cmd += ["-quiet"] # suppress errors about no 64-bit libs
run_drconfig = True
if self._options.follow_python:"Following python children")
# -unreg fails if not already registered so query for that first
query_cmd = drconfig_cmd + ["-isreg", "python.exe"]
query_proc = subprocess.Popen(query_cmd, stdout=subprocess.PIPE,
(query_out, query_err) = query_proc.communicate()
if"exe not registered", query_out):
run_drconfig = False # all set
drconfig_cmd += ["-unreg", "python.exe"]
else:"Excluding python children")
drconfig_cmd += ["-reg", "python.exe", "-norun"]
if run_drconfig:
drconfig_retcode = common.RunSubprocess(drconfig_cmd, self._timeout)
if drconfig_retcode:
logging.error("Configuring whether to follow python children failed " \
"with %d.", drconfig_retcode)
raise RuntimeError, "Configuring python children failed "
suppression_count = 0
for suppression_file in self._options.suppressions:
if os.path.exists(suppression_file):
suppression_count += 1
proc += ["-suppress", suppression_file]
if not suppression_count:
# Un-comment to dump Dr.Memory events on error
#proc += ["-dr_ops", "-dumpcore_mask 0x8bff"]
# Un-comment to debug Dr.Memory
#proc += ["-dr_ops", "-no_hide -msgbox_mask 15"]
if self._options.use_debug:
proc += ["-debug"]
proc += ["-logdir", self.log_dir]
proc += ["-batch", "-quiet", "-no_results_to_stderr"]
proc += ["-callstack_max_frames", "40"]
# make callstacks easier to read
proc += ["-callstack_srcfile_prefix",
proc += ["-callstack_modname_hide",
boring_callers = common.BoringCallers(mangled=False, use_re_wildcards=False)
# TODO(timurrrr): In fact, we want "starting from .." instead of "below .."
proc += ["-callstack_truncate_below", ",".join(boring_callers)]
if not self.handle_uninits_and_leaks:
proc += ["-no_check_uninitialized", "-no_count_leaks"]
proc += self._tool_flags
# Dr.Memory requires -- to separate tool flags from the executable name.
proc += ["--"]
if self._options.indirect:
self.CreateBrowserWrapper(" ".join(proc))
proc = []
# Note that self._args begins with the name of the exe to be run.
proc += self._args
return proc
def CreateBrowserWrapper(self, command):
os.putenv("BROWSER_WRAPPER", command)
def Analyze(self, check_sanity=False):
# Glob all the results files in the "testing.tmp" directory
filenames = glob.glob(self.log_dir + "/*/results.txt")
analyzer = drmemory_analyze.DrMemoryAnalyze(self._source_dir, filenames)
ret = analyzer.Report(check_sanity)
return ret
# RaceVerifier support. See
# for more details.
class ThreadSanitizerRV1Analyzer(tsan_analyze.TsanAnalyzer):
""" TsanAnalyzer that saves race reports to a file. """
TMP_FILE = "rvlog.tmp"
def __init__(self, source_dir, use_gdb):
super(ThreadSanitizerRV1Analyzer, self).__init__(source_dir, use_gdb)
self.out = open(self.TMP_FILE, "w")
def Report(self, files, check_sanity=False):
reports = self.GetReports(files)
for report in reports:
print >>self.out, report
if len(reports) > 0:"RaceVerifier pass 1 of 2, found %i reports" % len(reports))
return -1
return 0
def CloseOutputFile(self):
class ThreadSanitizerRV1Mixin(object):
"""RaceVerifier first pass.
Runs ThreadSanitizer as usual, but hides race reports and collects them in a
temporary file"""
def __init__(self):
super(ThreadSanitizerRV1Mixin, self).__init__()
def ExtendOptionParser(self, parser):
def CreateAnalyzer(self):
use_gdb = common.IsMac()
self.analyzer = ThreadSanitizerRV1Analyzer(self._source_dir, use_gdb)
return self.analyzer
def Cleanup(self):
super(ThreadSanitizerRV1Mixin, self).Cleanup()
class ThreadSanitizerRV2Mixin(object):
"""RaceVerifier second pass."""
def __init__(self):
super(ThreadSanitizerRV2Mixin, self).__init__()
def ExtendOptionParser(self, parser):
parser.add_option("", "--race-verifier-sleep-ms",
dest="race_verifier_sleep_ms", default=10,
help="duration of RaceVerifier delays")
def ToolSpecificFlags(self):
proc = super(ThreadSanitizerRV2Mixin, self).ToolSpecificFlags()
proc += ['--race-verifier=%s' % ThreadSanitizerRV1Analyzer.TMP_FILE,
'--race-verifier-sleep-ms=%d' %
return proc
def Cleanup(self):
super(ThreadSanitizerRV2Mixin, self).Cleanup()
class ThreadSanitizerRV1Posix(ThreadSanitizerRV1Mixin, ThreadSanitizerPosix):
class ThreadSanitizerRV2Posix(ThreadSanitizerRV2Mixin, ThreadSanitizerPosix):
class ThreadSanitizerRV1Windows(ThreadSanitizerRV1Mixin,
class ThreadSanitizerRV2Windows(ThreadSanitizerRV2Mixin,
class RaceVerifier(object):
"""Runs tests under RaceVerifier/Valgrind."""
def RV1Factory(self):
if common.IsWindows():
return ThreadSanitizerRV1Windows()
return ThreadSanitizerRV1Posix()
def RV2Factory(self):
if common.IsWindows():
return ThreadSanitizerRV2Windows()
return ThreadSanitizerRV2Posix()
def ToolName(self):
return "tsan"
def Main(self, args, check_sanity):"Running a TSan + RaceVerifier test. For more information, " +
"see " + self.MORE_INFO_URL)
cmd1 = self.RV1Factory()
ret = cmd1.Main(args, check_sanity)
# Verify race reports, if there are any.
if ret == -1:"Starting pass 2 of 2. Running the same binary in " +
"RaceVerifier mode to confirm possible race reports.")"For more information, see " + self.MORE_INFO_URL)
cmd2 = self.RV2Factory()
ret = cmd2.Main(args, check_sanity)
else:"No reports, skipping RaceVerifier second pass")"Please see " + self.MORE_INFO_URL + " for more information " +
"on RaceVerifier")
return ret
def Run(self, args, module):
return self.Main(args, False)
class EmbeddedTool(BaseTool):
"""Abstract class for tools embedded directly into the test binary.
# TODO(glider): need to override Execute() and support process chaining here.
def ToolCommand(self):
# In the simplest case just the args of the script.
return self._args
class Asan(EmbeddedTool):
"""AddressSanitizer, a memory error detector.
More information at
def __init__(self):
super(Asan, self).__init__()
self._timeout = 1200
if common.IsMac():
self._env["DYLD_NO_PIE"] = "1"
def ToolName(self):
return "asan"
def ToolCommand(self):
# TODO(glider): use pipes instead of the ugly wrapper here once they
# are supported.
procs = [os.path.join(self._source_dir, "tools", "valgrind",
"asan", "")]
return procs
def Analyze(sels, unused_check_sanity):
return 0
class TsanGcc(EmbeddedTool):
"""ThreadSanitizer with compile-time instrumentation done using GCC.
More information at
def __init__(self):
super(TsanGcc, self).__init__()
def ExtendOptionParser(self, parser):
parser.add_option("", "--suppressions", default=[],
help="path to TSan suppression file")
def Setup(self, args):
if not super(TsanGcc, self).Setup(args):
return False
ld_library_paths = []
for tail in "lib32", "lib64":
ld_library_paths.append(os.path.join(self._source_dir, "third_party",
"compiler-tsan", "gcc-4.5.3", tail))
# LD_LIBRARY_PATH will be overriden.
self._env["LD_LIBRARY_PATH"] = ":".join(ld_library_paths)
# TODO(glider): this is a temporary solution until Analyze is implemented.
env_options = ["--error-exitcode=1"]
# TODO(glider): merge this with other TSan suppressions code.
suppression_count = 0
for suppression_file in self._options.suppressions:
if os.path.exists(suppression_file):
suppression_count += 1
env_options += ["--suppressions=%s" % suppression_file]
if not suppression_count:
self._env["TSAN_ARGS"] = " ".join(env_options)
return True
def ToolName(self):
return "tsan"
def Analyze(self, unused_check_sanity):
# TODO(glider): this should use tsan_analyze.TsanAnalyzer. As a temporary
# solution we set the exit code to 1 when a report occurs, because TSan-GCC
# does not support the --log-file flag yet.
return 0
class ToolFactory:
def Create(self, tool_name):
if tool_name == "memcheck":
return Memcheck()
if tool_name == "tsan":
if common.IsWindows():
return ThreadSanitizerWindows()
return ThreadSanitizerPosix()
if tool_name == "drmemory" or tool_name == "drmemory_light":
# TODO(timurrrr): remove support for "drmemory" when buildbots are
# switched to drmemory_light OR make drmemory==drmemory_full the default
# mode when the tool is mature enough.
return DrMemory(False)
if tool_name == "drmemory_full":
return DrMemory(True)
if tool_name == "tsan_rv":
return RaceVerifier()
if tool_name == "tsan_gcc":
return TsanGcc()
if tool_name == "asan":
return Asan()
platform_name = common.PlatformNames()[0]
except common.NotImplementedError:
platform_name = sys.platform + "(Unknown)"
raise RuntimeError, "Unknown tool (tool=%s, platform=%s)" % (tool_name,
def CreateTool(tool):
return ToolFactory().Create(tool)
if __name__ == '__main__':
logging.error(sys.argv[0] + " can not be run from command line")