提交 dce45cf8 authored 作者: Frédéric Bastien's avatar Frédéric Bastien

Merge pull request #2310 from abergeron/compile_time

Improve compile time on slow filesystems
...@@ -166,7 +166,7 @@ import theano and print the config variable, as in: ...@@ -166,7 +166,7 @@ import theano and print the config variable, as in:
Theano initialize the GPU device. Newer version of PyCUDA Theano initialize the GPU device. Newer version of PyCUDA
(currently only in the trunk) don't have this restriction. (currently only in the trunk) don't have this restriction.
.. attribute:: config.print_active_device .. attribute:: print_active_device
Bool value: either ``True`` or ``False`` Bool value: either ``True`` or ``False``
...@@ -310,26 +310,26 @@ import theano and print the config variable, as in: ...@@ -310,26 +310,26 @@ import theano and print the config variable, as in:
Do the vm/cvm linkers profile the optimization phase when compiling a Theano function? Do the vm/cvm linkers profile the optimization phase when compiling a Theano function?
It only works when profile=True. It only works when profile=True.
.. attribute:: profiling.n_apply .. attribute:: config.profiling.n_apply
Positive int value, default: 20. Positive int value, default: 20.
The number of Apply nodes to print in the profiler output The number of Apply nodes to print in the profiler output
.. attribute:: profiling.n_ops .. attribute:: config.profiling.n_ops
Positive int value, default: 20. Positive int value, default: 20.
The number of Ops to print in the profiler output The number of Ops to print in the profiler output
.. attribute:: profiling.min_memory_size .. attribute:: config.profiling.min_memory_size
Positive int value, default: 1024. Positive int value, default: 1024.
For the memory profile, do not print Apply nodes if the size For the memory profile, do not print Apply nodes if the size
of their outputs (in bytes) is lower than this. of their outputs (in bytes) is lower than this.
.. attribute:: profiling.min_peak_memory .. attribute:: config.profiling.min_peak_memory
Bool value: either True or False Bool value: either True or False
...@@ -338,7 +338,7 @@ import theano and print the config variable, as in: ...@@ -338,7 +338,7 @@ import theano and print the config variable, as in:
Does the memory profile print the min peak memory usage? Does the memory profile print the min peak memory usage?
It only works when profile=True, profile_memory=True It only works when profile=True, profile_memory=True
.. attribute:: profiling.destination .. attribute:: config.profiling.destination
String value: 'stderr', 'stdout', or a name of a String value: 'stderr', 'stdout', or a name of a
file to be created file to be created
...@@ -491,7 +491,7 @@ import theano and print the config variable, as in: ...@@ -491,7 +491,7 @@ import theano and print the config variable, as in:
Extra parameters to pass to gcc when compiling. Extra include paths, Extra parameters to pass to gcc when compiling. Extra include paths,
library paths, configuration options, etc. library paths, configuration options, etc.
.. attribute:: config.cxx .. attribute:: cxx
Default: 'g++' if g++ is present. Empty string otherwise. Default: 'g++' if g++ is present. Empty string otherwise.
...@@ -555,7 +555,33 @@ import theano and print the config variable, as in: ...@@ -555,7 +555,33 @@ import theano and print the config variable, as in:
This means files whose compilation failed are deleted. This means files whose compilation failed are deleted.
Set to True to keep those files in order to debug compilation errors. Set to True to keep those files in order to debug compilation errors.
.. attribute:: config.DebugMode .. attribute:: compile
This section contains attributes which influence the compilation of
C code for ops. Due to historical reasons many attributes outside
of this section also have an influence over compilation, most
notably 'cxx'. This is not expected to change any time soon.
.. attribute:: config.compile.timeout
Positive int value, default: :attr:`compile.wait` * 24
Time to wait before an unrefreshed lock is broken and stolen. This
is in place to avoid manual cleanup of locks in case a process
crashed and left a lock in place.
The refresh time is automatically set to half the timeout value.
.. attribute:: config.compile.wait
Positive int value, default: 5
Time to wait between attempts at grabbing the lock if the first
attempt is not successful. The actual time will be between
:attr:`compile.wait` and :attr:`compile.wait` * 2 to avoid a
crowding effect on lock.
.. attribute:: DebugMode
This section contains various attributes configuring the behaviour This section contains various attributes configuring the behaviour
of mode :class:`~debugmode.DebugMode`. See directly this section of mode :class:`~debugmode.DebugMode`. See directly this section
...@@ -595,7 +621,7 @@ import theano and print the config variable, as in: ...@@ -595,7 +621,7 @@ import theano and print the config variable, as in:
Generate a warning when the destroy_map or view_map tell that an op work Generate a warning when the destroy_map or view_map tell that an op work
inplace, but the op did not reuse the input for its output. inplace, but the op did not reuse the input for its output.
.. attribute:: config.numpy .. attribute:: numpy
This section contains different attributes for configuring numpy's This section contains different attributes for configuring numpy's
behaviour, described by `numpy.seterr behaviour, described by `numpy.seterr
...@@ -669,7 +695,7 @@ import theano and print the config variable, as in: ...@@ -669,7 +695,7 @@ import theano and print the config variable, as in:
This flag's value cannot be modified during the program execution. This flag's value cannot be modified during the program execution.
.. attribute:: config.compute_test_value .. attribute:: compute_test_value
String Value: ``'off'``, ``'ignore'``, ``'warn'``, ``'raise'``. String Value: ``'off'``, ``'ignore'``, ``'warn'``, ``'raise'``.
...@@ -693,13 +719,13 @@ import theano and print the config variable, as in: ...@@ -693,13 +719,13 @@ import theano and print the config variable, as in:
this Op this Op
- ``'raise'`` will raise an Exception - ``'raise'`` will raise an Exception
.. attribute:: config.compute_test_value_opt .. attribute:: compute_test_value_opt
As ``compute_test_value``, but it is the value used during Theano As ``compute_test_value``, but it is the value used during Theano
optimization phase. Theano user's do not need to use this. This is optimization phase. Theano user's do not need to use this. This is
to help debug shape error in Theano optimization. to help debug shape error in Theano optimization.
.. attribute:: config.reoptimize_unpickled_function .. attribute:: reoptimize_unpickled_function
Bool value, default: True Bool value, default: True
...@@ -709,7 +735,7 @@ import theano and print the config variable, as in: ...@@ -709,7 +735,7 @@ import theano and print the config variable, as in:
reoptimized when being unpickled. Otherwise, skip the graph optimization and reoptimized when being unpickled. Otherwise, skip the graph optimization and
use directly the optimized graph. use directly the optimized graph.
.. attribute:: config.exception_verbosity .. attribute:: exception_verbosity
String Value: ``'low'``, ``'high'``. String Value: ``'low'``, ``'high'``.
...@@ -749,13 +775,13 @@ import theano and print the config variable, as in: ...@@ -749,13 +775,13 @@ import theano and print the config variable, as in:
This is useful to debug in gdb modules compiled by Theano. This is useful to debug in gdb modules compiled by Theano.
The parameter -g is passed by default to g++. The parameter -g is passed by default to g++.
.. attribute:: cmodule.compilation_warning .. attribute:: config.cmodule.compilation_warning
Bool value, default: False Bool value, default: False
If True, will print compilation warnings. If True, will print compilation warnings.
.. attribute:: cmodule.preload_cache' .. attribute:: config.cmodule.preload_cache
Bool value, default: False Bool value, default: False
......
...@@ -132,29 +132,38 @@ AddConfigVar('mode', ...@@ -132,29 +132,38 @@ AddConfigVar('mode',
'FAST_COMPILE', 'PROFILE_MODE', 'DEBUG_MODE'), 'FAST_COMPILE', 'PROFILE_MODE', 'DEBUG_MODE'),
in_c_key=False) in_c_key=False)
param = StrParam("g++") param = "g++"
# Test whether or not g++ is present: disable C code if it is not. # Test whether or not g++ is present: disable C code if it is not.
try: try:
rc = call_subprocess_Popen(['g++', '-v']) rc = call_subprocess_Popen(['g++', '-v'])
except OSError: except OSError:
param = StrParam("") param = ""
rc = 1 rc = 1
# On Mac we test for 'clang++' and use it by default # On Mac we test for 'clang++' and use it by default
if sys.platform == 'darwin': if sys.platform == 'darwin':
try: try:
rc = call_subprocess_Popen(['clang++', '-v']) rc = call_subprocess_Popen(['clang++', '-v'])
param = StrParam("clang++") param = "clang++"
except OSError: except OSError:
pass pass
# Try to find the full compiler path from the name
if param != "":
import distutils.spawn
newp = distutils.spawn.find_executable(param)
if newp is not None:
param = newp
del newp
del distutils
AddConfigVar('cxx', AddConfigVar('cxx',
"The C++ compiler to use. Currently only g++ is" "The C++ compiler to use. Currently only g++ is"
" supported, but supporting additional compilers should not be " " supported, but supporting additional compilers should not be "
"too difficult. " "too difficult. "
"If it is empty, no C++ code is compiled.", "If it is empty, no C++ code is compiled.",
param, StrParam(param),
in_c_key=False) in_c_key=False)
del param del param
......
...@@ -9,38 +9,39 @@ import time ...@@ -9,38 +9,39 @@ import time
import logging import logging
from theano import config from theano import config
from theano.configparser import AddConfigVar, IntParam
_logger = logging.getLogger("theano.gof.compilelock") _logger = logging.getLogger("theano.gof.compilelock")
# If the user provided a logging level, we don't want to override it. # If the user provided a logging level, we don't want to override it.
if _logger.level == logging.NOTSET: if _logger.level == logging.NOTSET:
# INFO will show the the messages "Refreshing lock" message # INFO will show the "Refreshing lock" messages
_logger.setLevel(logging.INFO) _logger.setLevel(logging.INFO)
# In seconds, time that a process will wait before deciding to override an AddConfigVar('compile.wait',
# existing lock. An override only happens when the existing lock is held by """Time to wait before retrying to aquire the compile lock.""",
# the same owner *and* has not been 'refreshed' by this owner for more than IntParam(5, lambda i: i > 0, allow_override=False),
# 'timeout_before_override' seconds. in_c_key=False)
timeout_before_override = 120
# In seconds, duration before a lock is refreshed. More precisely, the lock is def _timeout_default():
# refreshed each time 'get_lock()' is called (typically for each file being return config.compile.wait * 24
# compiled) and the existing lock has not been refreshed in the past
# 'refresh_every' seconds. AddConfigVar('compile.timeout',
refresh_every = 60 """In seconds, time that a process will wait before deciding to
override an existing lock. An override only happens when the existing
lock is held by the same owner *and* has not been 'refreshed' by this
owner for more than this period. Refreshes are done every half timeout
period for running processes.""",
IntParam(_timeout_default, lambda i: i >= 0,
allow_override=False),
in_c_key=False)
def force_unlock(): def force_unlock():
""" """
Delete the compilation lock if someone else has it. Delete the compilation lock if someone else has it.
""" """
global timeout_before_override get_lock(min_wait=0, max_wait=0.001, timeout=0)
timeout_backup = timeout_before_override release_lock()
timeout_before_override = 0
try:
get_lock(min_wait=0, max_wait=0.001)
release_lock()
finally:
timeout_before_override = timeout_backup
def get_lock(lock_dir=None, **kw): def get_lock(lock_dir=None, **kw):
...@@ -74,16 +75,17 @@ def get_lock(lock_dir=None, **kw): ...@@ -74,16 +75,17 @@ def get_lock(lock_dir=None, **kw):
if get_lock.lock_is_enabled: if get_lock.lock_is_enabled:
# Only really try to acquire the lock if we do not have it already. # Only really try to acquire the lock if we do not have it already.
if get_lock.n_lock == 0: if get_lock.n_lock == 0:
lock(get_lock.lock_dir, timeout=timeout_before_override, **kw) lock(get_lock.lock_dir, **kw)
atexit.register(Unlocker.unlock, get_lock.unlocker) atexit.register(Unlocker.unlock, get_lock.unlocker)
# Store time at which the lock was set. # Store time at which the lock was set.
get_lock.start_time = time.time() get_lock.start_time = time.time()
else: else:
# Check whether we need to 'refresh' the lock. We do this every # Check whether we need to 'refresh' the lock. We do this
# 'refresh_every' seconds to ensure noone else tries to override # every 'config.compile.timeout / 2' seconds to ensure
# our lock after their 'timeout_before_override' timeout period. # no one else tries to override our lock after their
# 'config.compile.timeout' timeout period.
now = time.time() now = time.time()
if now - get_lock.start_time > refresh_every: if now - get_lock.start_time > config.compile.timeout/2:
lockpath = os.path.join(get_lock.lock_dir, 'lock') lockpath = os.path.join(get_lock.lock_dir, 'lock')
_logger.info('Refreshing lock %s', str(lockpath)) _logger.info('Refreshing lock %s', str(lockpath))
refresh_lock(lockpath) refresh_lock(lockpath)
...@@ -109,46 +111,57 @@ def set_lock_status(use_lock): ...@@ -109,46 +111,57 @@ def set_lock_status(use_lock):
by default). Disabling may make compilation slightly faster (but is not by default). Disabling may make compilation slightly faster (but is not
recommended for parallel execution). recommended for parallel execution).
@param use_lock: whether to use the compilation lock or not :param use_lock: whether to use the compilation lock or not
@type use_lock: bool :type use_lock: bool
""" """
get_lock.lock_is_enabled = use_lock get_lock.lock_is_enabled = use_lock
# This is because None is a valid input for timeout
notset = object()
def lock(tmp_dir, timeout=120, min_wait=5, max_wait=10, verbosity=1): def lock(tmp_dir, timeout=notset, min_wait=None, max_wait=None, verbosity=1):
""" """
Obtain lock access by creating a given temporary directory (whose base will Obtain lock access by creating a given temporary directory (whose base will
be created if needed, but will not be deleted after the lock is removed). be created if needed, but will not be deleted after the lock is removed).
If access is refused by the same lock owner during more than 'timeout' If access is refused by the same lock owner during more than 'timeout'
seconds, then the current lock is overridden. If timeout is None, then no seconds, then the current lock is overridden. If timeout is None, then no
timeout is performed. timeout is performed.
The lock is performed by creating a 'lock' file in 'tmp_dir' that contains The lock is performed by creating a 'lock' file in 'tmp_dir' that contains
a unique id identifying the owner of the lock (the process id, followed by a unique id identifying the owner of the lock (the process id, followed by
a random string). a random string).
When there is already a lock, the process sleeps for a random amount of When there is already a lock, the process sleeps for a random amount of
time between min_wait and max_wait seconds before trying again. time between min_wait and max_wait seconds before trying again.
If 'verbosity' is >= 1, then a message will be displayed when we need to If 'verbosity' is >= 1, then a message will be displayed when we need to
wait for the lock. If it is set to a value >1, then this message will be wait for the lock. If it is set to a value >1, then this message will be
displayed each time we re-check for the presence of the lock. Otherwise it displayed each time we re-check for the presence of the lock. Otherwise it
is displayed only when we notice the lock's owner has changed. is displayed only when we notice the lock's owner has changed.
@param tmp_dir: lock directory that will be created when acquiring the lock :param str tmp_dir: lock directory that will be created when
@type tmp_dir: string acquiring the lock
@param timeout: time (in seconds) to wait before replacing an existing lock :param timeout: time (in seconds) to wait before replacing an
@type timeout: int or None existing lock (default config 'compile.timeout')
:type timeout: int or None
@param min_wait: minimum time (in seconds) to wait before trying again to :param int min_wait: minimum time (in seconds) to wait before
get the lock trying again to get the lock
@type min_wait: int (default config 'compile.wait')
@param max_wait: maximum time (in seconds) to wait before trying again to :param int max_wait: maximum time (in seconds) to wait before
get the lock trying again to get the lock
@type max_wait: int (default 2 * min_wait)
@param verbosity: amount of feedback displayed to screen :param int verbosity: amount of feedback displayed to screen (default 1)
@type verbosity: int
""" """
if min_wait is None:
min_wait = config.compile.wait
if max_wait is None:
max_wait = min_wait * 2
if timeout is notset:
timeout = config.compile.timeout
# Create base of lock directory if required. # Create base of lock directory if required.
base_lock = os.path.dirname(tmp_dir) base_lock = os.path.dirname(tmp_dir)
if not os.path.isdir(base_lock): if not os.path.isdir(base_lock):
...@@ -207,7 +220,7 @@ def lock(tmp_dir, timeout=120, min_wait=5, max_wait=10, verbosity=1): ...@@ -207,7 +220,7 @@ def lock(tmp_dir, timeout=120, min_wait=5, max_wait=10, verbosity=1):
continue continue
if last_owner == read_owner: if last_owner == read_owner:
if (timeout is not None and if (timeout is not None and
time.time() - time_start >= timeout): time.time() - time_start >= timeout):
# Timeout exceeded or locking process dead. # Timeout exceeded or locking process dead.
if not no_display: if not no_display:
if read_owner == 'failure': if read_owner == 'failure':
......
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论