view hgext/record.py @ 38732:be4984261611

merge: mark file gets as not thread safe (issue5933)

In default installs, this has the effect of disabling the thread-based
worker on Windows when manifesting files in the working directory. My
measurements have shown that with revlog-based repositories, Mercurial
spends a lot of CPU time in revlog code resolving file data. This ends
up incurring a lot of context switching across threads and slows down
`hg update` operations when going from an empty working directory to
the tip of the repo.

On mozilla-unified (246,351 files) on an i7-6700K (4+4 CPUs):

before: 487s wall
after:  360s wall (equivalent to worker.enabled=false)
cpus=2: 379s wall

Even with only 2 threads, the thread pool is still slower.

The introduction of the thread-based worker (02b36e860e0b) states that
it resulted in a "~50%" speedup for `hg sparse --enable-profile` and
`hg sparse --disable-profile`. This disagrees with my measurement
above. I theorize a few reasons for this:

1) Removal of files from the working directory is I/O - not CPU - bound
   and should benefit from a thread pool (unless I/O is insanely fast
   and the GIL release is near instantaneous). So tests like `hg sparse
   --enable-profile` may exercise deletion throughput and aren't good
   benchmarks for worker tasks that are CPU heavy.
2) The patch was authored by someone at Facebook. The results were
   likely measured against a repository using remotefilelog. And I
   believe that revision retrieval during working directory updates with
   remotefilelog will often use a remote store, thus being I/O and not
   CPU bound. This probably resulted in an overstated performance gain.

Since there appears to be a need to enable the thread-based worker with
some stores, I've made the flagging of file gets as thread safe
configurable. I've made it experimental because I don't want to formalize
a boolean flag for this option and because this attribute is best
captured against the store implementation. But we don't have a proper
store API for this yet. I'd rather cross this bridge later.

It is possible there are revlog-based repositories that do benefit from
a thread-based worker. I didn't do very comprehensive testing. If there
are, we may want to devise a more proper algorithm for whether to use
the thread-based worker, including possibly config options to limit the
number of threads to use. But until I see evidence that justifies
complexity, simplicity wins.

Differential Revision: https://phab.mercurial-scm.org/D3963

# record.py
#
# Copyright 2007 Bryan O'Sullivan <[e-mail]>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

'''commands to interactively select changes for commit/qrefresh (DEPRECATED)

The feature provided by this extension has been moved into core Mercurial as
:hg:`commit --interactive`.'''

from __future__ import absolute_import

from mercurial.i18n import _
from mercurial import (
    cmdutil,
    commands,
    error,
    extensions,
    registrar,
)

cmdtable = {}
command = registrar.command(cmdtable)
# Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
# extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
# be specifying the version(s) of Mercurial they are tested with, or
# leave the attribute unspecified.
testedwith = 'ships-with-hg-core'


@command("record",
         # same options as commit + white space diff options
        [c for c in commands.table['^commit|ci'][1][:]
            if c[1] != "interactive"] + cmdutil.diffwsopts,
          _('hg record [OPTION]... [FILE]...'))
def record(ui, repo, *pats, **opts):
    '''interactively select changes to commit

    If a list of files is omitted, all changes reported by :hg:`status`
    will be candidates for recording.

    See :hg:`help dates` for a list of formats valid for -d/--date.

    If using the text interface (see :hg:`help config`),
    you will be prompted for whether to record changes to each
    modified file, and for files with multiple changes, for each
    change to use. For each query, the following responses are
    possible::

      y - record this change
      n - skip this change
      e - edit this change manually

      s - skip remaining changes to this file
      f - record remaining changes to this file

      d - done, skip remaining changes and files
      a - record all changes to all remaining files
      q - quit, recording no changes

      ? - display help

    This command is not available when committing a merge.'''

    if not ui.interactive():
        raise error.Abort(_('running non-interactively, use %s instead') %
                         'commit')

    opts[r"interactive"] = True
    overrides = {('experimental', 'crecord'): False}
    with ui.configoverride(overrides, 'record'):
        return commands.commit(ui, repo, *pats, **opts)

def qrefresh(origfn, ui, repo, *pats, **opts):
    if not opts[r'interactive']:
        return origfn(ui, repo, *pats, **opts)

    mq = extensions.find('mq')

    def committomq(ui, repo, *pats, **opts):
        # At this point the working copy contains only changes that
        # were accepted. All other changes were reverted.
        # We can't pass *pats here since qrefresh will undo all other
        # changed files in the patch that aren't in pats.
        mq.refresh(ui, repo, **opts)

    # backup all changed files
    cmdutil.dorecord(ui, repo, committomq, None, True,
                    cmdutil.recordfilter, *pats, **opts)

# This command registration is replaced during uisetup().
@command('qrecord',
    [],
    _('hg qrecord [OPTION]... PATCH [FILE]...'),
    inferrepo=True)
def qrecord(ui, repo, patch, *pats, **opts):
    '''interactively record a new patch

    See :hg:`help qnew` & :hg:`help record` for more information and
    usage.
    '''
    return _qrecord('qnew', ui, repo, patch, *pats, **opts)

def _qrecord(cmdsuggest, ui, repo, patch, *pats, **opts):
    try:
        mq = extensions.find('mq')
    except KeyError:
        raise error.Abort(_("'mq' extension not loaded"))

    repo.mq.checkpatchname(patch)

    def committomq(ui, repo, *pats, **opts):
        opts[r'checkname'] = False
        mq.new(ui, repo, patch, *pats, **opts)

    overrides = {('experimental', 'crecord'): False}
    with ui.configoverride(overrides, 'record'):
        cmdutil.dorecord(ui, repo, committomq, cmdsuggest, False,
                         cmdutil.recordfilter, *pats, **opts)

def qnew(origfn, ui, repo, patch, *args, **opts):
    if opts[r'interactive']:
        return _qrecord(None, ui, repo, patch, *args, **opts)
    return origfn(ui, repo, patch, *args, **opts)


def uisetup(ui):
    try:
        mq = extensions.find('mq')
    except KeyError:
        return

    cmdtable["qrecord"] = \
        (qrecord,
         # same options as qnew, but copy them so we don't get
         # -i/--interactive for qrecord and add white space diff options
         mq.cmdtable['^qnew'][1][:] + cmdutil.diffwsopts,
         _('hg qrecord [OPTION]... PATCH [FILE]...'))

    _wrapcmd('qnew', mq.cmdtable, qnew, _("interactively record a new patch"))
    _wrapcmd('qrefresh', mq.cmdtable, qrefresh,
             _("interactively select changes to refresh"))

def _wrapcmd(cmd, table, wrapfn, msg):
    entry = extensions.wrapcommand(table, cmd, wrapfn)
    entry[1].append(('i', 'interactive', None, msg))