view tests/bruterebase.py @ 47343:9f798c1b0d89 stable

cext: fix memory leak in phases computation

Without this a buffer whose size in bytes is the number of
changesets in the repository is leaked each time the repository is
opened and changeset phases are computed.

Impact: the current code in hgwebdir creates a new `localrepository`
instance for each HTTP request. Since any pull or push is made of several
requests, a team of 100 people can easily produce thousands of such
requests per day.

Being a low-level malloc, this leak can't be seen with the gc module and
tools relying on that, but was spotted by valgrind immediately.

Reproduction
------------

  for i in range(cl_args.iterations):
      repo = hg.repository(baseui, repo_path)
      rev = repo.revs(rev).first()
      ctx = repo[rev]

  del ctx
  del repo
  # avoid any pollution by other type of leak
  # (that should be fixed in 5.8)
  repoview._filteredrepotypes.clear()

  gc.collect()

Measurements
------------

Resident Set Size (RSS), taken on a clone of
mozilla-central for performance analysis (440 000
changesets).

before:
  5.8+hg19.5ac0f2a8ba72  1000 iterations: 1606MB
  5.8+hg19.5ac0f2a8ba72 10000 iterations: 5723MB
after:
  5.8+hg20.e2084d39e145  1000 iterations:  555MB
  5.8+hg20.e2084d39e145 10000 iterations:  555MB
         (double checked, not a copy/paste error)

(e2084d39e14 is the present changeset, before amendment
of the message to add the measurements)

# bruterebase.py - brute force rebase testing
#
# Copyright 2017 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

from mercurial import (
    error,
    registrar,
    revsetlang,
)

from hgext import rebase

try:
    xrange
except NameError:
    xrange = range

cmdtable = {}
command = registrar.command(cmdtable)


@command(b'debugbruterebase')
def debugbruterebase(ui, repo, source, dest):
    """for every non-empty subset of source, run rebase -r subset -d dest

    Print one line summary for each subset. Assume obsstore is enabled.
    """
    srevs = list(repo.revs(source))

    with repo.wlock(), repo.lock():
        repolen = len(repo)
        cl = repo.changelog

        def getdesc(rev):
            result = cl.changelogrevision(rev).description
            if rev >= repolen:
                result += b"'"
            return result

        for i in xrange(1, 2 ** len(srevs)):
            subset = [rev for j, rev in enumerate(srevs) if i & (1 << j) != 0]
            spec = revsetlang.formatspec(b'%ld', subset)
            tr = repo.transaction(b'rebase')
            tr._report = lambda x: 0  # hide "transaction abort"

            ui.pushbuffer()
            try:
                rebase.rebase(ui, repo, dest=dest, rev=[spec])
            except error.Abort as ex:
                summary = b'ABORT: %s' % ex.message
            except Exception as ex:
                summary = b'CRASH: %s' % ex
            else:
                # short summary about new nodes
                cl = repo.changelog
                descs = []
                for rev in xrange(repolen, len(repo)):
                    desc = b'%s:' % getdesc(rev)
                    for prev in cl.parentrevs(rev):
                        if prev > -1:
                            desc += getdesc(prev)
                    descs.append(desc)
                descs.sort()
                summary = b' '.join(descs)
            ui.popbuffer()
            repo.vfs.tryunlink(b'rebasestate')

            subsetdesc = b''.join(getdesc(rev) for rev in subset)
            ui.write(b'%s: %s\n' % (subsetdesc.rjust(len(srevs)), summary))
            tr.abort()