view mercurial/worker.py @ 31793:69d8fcf20014

help: document bundle specifications

I softly formalized the concept of a "bundle specification" a while
ago when I was working on clone bundles and stream clone bundles and
wanted a more robust way to define what exactly is in a bundle file.

The concept has existed for a while. Since it is part of the clone
bundles feature and exposed to the user via the "-t" argument to
`hg bundle`, it is something we need to support for the long haul.

After the 4.1 release, I heard a few people comment that they didn't
realize you could generate zstd bundles with `hg bundle`. I'm
partially to blame for not documenting it in bundle's docstring.

Additionally, I added a hacky, experimental feature for controlling
the compression level of bundles in 76104a4899ad. As the commit
message says, I went with a quick and dirty solution out of time
constraints. Furthermore, I wanted to eventually store this
configuration in the "bundlespec" so it could be made more flexible.

Given:

a) bundlespecs are here to stay
b) we don't have great documentation over what they are, despite being
   a user-facing feature
c) the list of available compression engines and their behavior isn't
   exposed
d) we need an extensible place to modify behavior of compression
   engines

I want to move forward with formalizing bundlespecs as a user-facing
feature. This commit does that by introducing a "bundlespec" help
page. Leaning on the just-added compression engine documentation
and API, the topic also conveniently lists available compression
engines and details about them. This makes features like zstd
bundle compression more discoverable. e.g. you can now
`hg help -k zstd` and it lists the "bundlespec" topic.

# worker.py - master-slave parallelism support
#
# Copyright 2013 Facebook, Inc.
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

from __future__ import absolute_import

import errno
import os
import signal
import sys

from .i18n import _
from . import (
    encoding,
    error,
    pycompat,
    scmutil,
    util,
)

def countcpus():
    '''try to count the number of CPUs on the system'''

    # posix
    try:
        n = int(os.sysconf('SC_NPROCESSORS_ONLN'))
        if n > 0:
            return n
    except (AttributeError, ValueError):
        pass

    # windows
    try:
        n = int(encoding.environ['NUMBER_OF_PROCESSORS'])
        if n > 0:
            return n
    except (KeyError, ValueError):
        pass

    return 1

def _numworkers(ui):
    s = ui.config('worker', 'numcpus')
    if s:
        try:
            n = int(s)
            if n >= 1:
                return n
        except ValueError:
            raise error.Abort(_('number of cpus must be an integer'))
    return min(max(countcpus(), 4), 32)

if pycompat.osname == 'posix':
    _startupcost = 0.01
else:
    _startupcost = 1e30

def worthwhile(ui, costperop, nops):
    '''try to determine whether the benefit of multiple processes can
    outweigh the cost of starting them'''
    linear = costperop * nops
    workers = _numworkers(ui)
    benefit = linear - (_startupcost * workers + linear / workers)
    return benefit >= 0.15

def worker(ui, costperarg, func, staticargs, args):
    '''run a function, possibly in parallel in multiple worker
    processes.

    returns a progress iterator

    costperarg - cost of a single task

    func - function to run

    staticargs - arguments to pass to every invocation of the function

    args - arguments to split into chunks, to pass to individual
    workers
    '''
    if worthwhile(ui, costperarg, len(args)):
        return _platformworker(ui, func, staticargs, args)
    return func(*staticargs + (args,))

def _posixworker(ui, func, staticargs, args):
    rfd, wfd = os.pipe()
    workers = _numworkers(ui)
    oldhandler = signal.getsignal(signal.SIGINT)
    signal.signal(signal.SIGINT, signal.SIG_IGN)
    pids, problem = set(), [0]
    def killworkers():
        # unregister SIGCHLD handler as all children will be killed. This
        # function shouldn't be interrupted by another SIGCHLD; otherwise pids
        # could be updated while iterating, which would cause inconsistency.
        signal.signal(signal.SIGCHLD, oldchldhandler)
        # if one worker bails, there's no good reason to wait for the rest
        for p in pids:
            try:
                os.kill(p, signal.SIGTERM)
            except OSError as err:
                if err.errno != errno.ESRCH:
                    raise
    def waitforworkers(blocking=True):
        for pid in pids.copy():
            p = st = 0
            while True:
                try:
                    p, st = os.waitpid(pid, (0 if blocking else os.WNOHANG))
                    break
                except OSError as e:
                    if e.errno == errno.EINTR:
                        continue
                    elif e.errno == errno.ECHILD:
                        # child would already be reaped, but pids yet been
                        # updated (maybe interrupted just after waitpid)
                        pids.discard(pid)
                        break
                    else:
                        raise
            if not p:
                # skip subsequent steps, because child process should
                # be still running in this case
                continue
            pids.discard(p)
            st = _exitstatus(st)
            if st and not problem[0]:
                problem[0] = st
    def sigchldhandler(signum, frame):
        waitforworkers(blocking=False)
        if problem[0]:
            killworkers()
    oldchldhandler = signal.signal(signal.SIGCHLD, sigchldhandler)
    ui.flush()
    for pargs in partition(args, workers):
        pid = os.fork()
        if pid == 0:
            signal.signal(signal.SIGINT, oldhandler)
            signal.signal(signal.SIGCHLD, oldchldhandler)

            def workerfunc():
                os.close(rfd)
                for i, item in func(*(staticargs + (pargs,))):
                    os.write(wfd, '%d %s\n' % (i, item))

            # make sure we use os._exit in all code paths. otherwise the worker
            # may do some clean-ups which could cause surprises like deadlock.
            # see sshpeer.cleanup for example.
            try:
                try:
                    scmutil.callcatch(ui, workerfunc)
                finally:
                    ui.flush()
            except KeyboardInterrupt:
                os._exit(255)
            except: # never return, therefore no re-raises
                try:
                    ui.traceback()
                    ui.flush()
                finally:
                    os._exit(255)
            else:
                os._exit(0)
        pids.add(pid)
    os.close(wfd)
    fp = os.fdopen(rfd, pycompat.sysstr('rb'), 0)
    def cleanup():
        signal.signal(signal.SIGINT, oldhandler)
        waitforworkers()
        signal.signal(signal.SIGCHLD, oldchldhandler)
        status = problem[0]
        if status:
            if status < 0:
                os.kill(os.getpid(), -status)
            sys.exit(status)
    try:
        for line in util.iterfile(fp):
            l = line.split(' ', 1)
            yield int(l[0]), l[1][:-1]
    except: # re-raises
        killworkers()
        cleanup()
        raise
    cleanup()

def _posixexitstatus(code):
    '''convert a posix exit status into the same form returned by
    os.spawnv

    returns None if the process was stopped instead of exiting'''
    if os.WIFEXITED(code):
        return os.WEXITSTATUS(code)
    elif os.WIFSIGNALED(code):
        return -os.WTERMSIG(code)

if pycompat.osname != 'nt':
    _platformworker = _posixworker
    _exitstatus = _posixexitstatus

def partition(lst, nslices):
    '''partition a list into N slices of roughly equal size

    The current strategy takes every Nth element from the input. If
    we ever write workers that need to preserve grouping in input
    we should consider allowing callers to specify a partition strategy.

    mpm is not a fan of this partitioning strategy when files are involved.
    In his words:

        Single-threaded Mercurial makes a point of creating and visiting
        files in a fixed order (alphabetical). When creating files in order,
        a typical filesystem is likely to allocate them on nearby regions on
        disk. Thus, when revisiting in the same order, locality is maximized
        and various forms of OS and disk-level caching and read-ahead get a
        chance to work.

        This effect can be quite significant on spinning disks. I discovered it
        circa Mercurial v0.4 when revlogs were named by hashes of filenames.
        Tarring a repo and copying it to another disk effectively randomized
        the revlog ordering on disk by sorting the revlogs by hash and suddenly
        performance of my kernel checkout benchmark dropped by ~10x because the
        "working set" of sectors visited no longer fit in the drive's cache and
        the workload switched from streaming to random I/O.

        What we should really be doing is have workers read filenames from a
        ordered queue. This preserves locality and also keeps any worker from
        getting more than one file out of balance.
    '''
    for i in range(nslices):
        yield lst[i::nslices]