Mercurial > hg
view mercurial/subrepoutil.py @ 42044:bb271ec2fbfb
compression: introduce a `storage.revlog.zstd.level` configuration
This option control the zstd compression level used when compressing revlog
chunk. The usage of zstd for revlog compression has not graduated from
experimental yet, but we intend to fix that soon.
The option name for the compression level is more straight forward to pick, so
this changesets comes first. Having a dedicated option for each compression
engine is useful because they don't support the same range of values.
I ran the same measurement as for the zlib compression level (in the parent
changesets). The variation in repository size is stay mostly in the same (small)
range. The "read/write" performance see smallish variation, but are overall much
better than zlib. Write performance show the same tend of having better write
performance for when reaching high-end compression.
Again, we don't intend to change the default zstd compression level (currently:
3) in this series. However this is worth investigating in the future.
The Performance comparison of zlib vs zstd is quite impressive. The repository
size stay in the same range, but the performance are much better in all
situations.
Comparison summary
==================
We are looking at:
- performance range for zlib
- performance range for zstd
- comparison of default zstd (level-3) to default zlib (level 6)
- comparison of the slowest zstd time to the fastest zlib time
Read performance:
-----------------
| zlib | zstd | cmp | f2s
mercurial | 0.170159 - 0.189219 | 0.144127 - 0.149624 | 80% | 88%
pypy | 2.679217 - 2.768691 | 1.532317 - 1.705044 | 60% | 63%
netbeans | 122.477027 - 141.620281 | 72.996346 - 89.731560 | 58% | 73%
mozilla | 147.867662 - 170.572118 | 91.700995 - 105.853099 | 56% | 71%
Write performance:
------------------
| zlib | zstd | cmp | f2s
mercurial | 53.250304 - 56.2936129 | 40.877025 - 45.677286 | 75% | 86%
pypy | 460.721984 - 476.589918 | 270.545409 - 301.002219 | 63% | 65%
netbeans | 520.560316 - 715.930400 | 370.356311 - 428.329652 | 55% | 82%
mozilla | 739.803002 - 987.056093 | 505.152906 - 591.930683 | 57% | 80%
Raw data
--------
repo alg lvl .hg/store size 00manifest.d read write
mercurial zlib 1 49,402,813 5,963,475 0.170159 53.250304
mercurial zlib 6 47,197,397 5,875,730 0.182820 56.264320
mercurial zlib 9 47,121,596 5,849,781 0.189219 56.293612
mercurial zstd 1 49,737,084 5,966,355 0.144127 40.877025
mercurial zstd 3 48,961,867 5,895,208 0.146376 42.268142
mercurial zstd 5 48,200,592 5,938,676 0.149624 43.162875
mercurial zstd 10 47,833,520 5,913,353 0.145185 44.012489
mercurial zstd 15 47,314,604 5,728,679 0.147686 45.677286
mercurial zstd 20 47,330,502 5,830,539 0.145789 45.025407
mercurial zstd 22 47,330,076 5,830,539 0.143996 44.690460
pypy zlib 1 370,830,572 28,462,425 2.679217 460.721984
pypy zlib 6 340,112,317 27,648,747 2.768691 467.537158
pypy zlib 9 338,360,736 27,639,003 2.763495 476.589918
pypy zstd 1 362,377,479 27,916,214 1.532317 270.545409
pypy zstd 3 354,137,693 27,905,988 1.686718 294.951509
pypy zstd 5 342,640,043 27,655,774 1.705044 301.002219
pypy zstd 10 334,224,327 27,164,493 1.567287 285.186239
pypy zstd 15 329,000,363 26,645,965 1.637729 299.561332
pypy zstd 20 324,534,039 26,199,547 1.526813 302.149827
pypy zstd 22 324,530,595 26,198,932 1.525718 307.821218
netbeans zlib 1 1,281,847,810 165,495,457 122.477027 520.560316
netbeans zlib 6 1,205,284,353 159,161,207 139.876147 715.930400
netbeans zlib 9 1,197,135,671 155,034,586 141.620281 678.297064
netbeans zstd 1 1,259,581,737 160,840,613 72.996346 370.356311
netbeans zstd 3 1,232,978,122 157,691,551 81.622317 396.733087
netbeans zstd 5 1,208,034,075 160,246,880 83.080549 364.342626
netbeans zstd 10 1,188,624,176 156,083,417 79.323935 403.594602
netbeans zstd 15 1,176,973,589 153,859,477 89.731560 428.329652
netbeans zstd 20 1,162,958,258 151,147,535 82.842667 392.335349
netbeans zstd 22 1,162,707,029 151,150,220 82.565695 402.840655
mozilla zlib 1 2,775,497,186 298,527,987 147.867662 751.263721
mozilla zlib 6 2,596,856,420 286,597,671 170.572118 987.056093
mozilla zlib 9 2,587,542,494 287,018,264 163.622338 739.803002
mozilla zstd 1 2,723,159,348 286,617,532 91.700995 570.042751
mozilla zstd 3 2,665,055,001 286,152,013 95.240155 561.412805
mozilla zstd 5 2,607,819,817 288,060,030 101.978048 505.152906
mozilla zstd 10 2,558,761,085 283,967,648 104.113481 497.771202
mozilla zstd 15 2,526,216,060 275,581,300 105.853099 591.930683
mozilla zstd 20 2,485,114,806 266,478,859 95.268795 576.515389
mozilla zstd 22 2,484,869,080 266,456,505 94.429282 572.785537
author | Pierre-Yves David <pierre-yves.david@octobus.net> |
---|---|
date | Wed, 27 Mar 2019 18:35:59 +0100 |
parents | 876494fd967d |
children | f6540aba8e3e |
line wrap: on
line source
# subrepoutil.py - sub-repository operations and substate handling # # Copyright 2009-2010 Matt Mackall <mpm@selenic.com> # # 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 posixpath import re from .i18n import _ from . import ( config, error, filemerge, pathutil, phases, util, ) from .utils import ( stringutil, ) nullstate = ('', '', 'empty') def state(ctx, ui): """return a state dict, mapping subrepo paths configured in .hgsub to tuple: (source from .hgsub, revision from .hgsubstate, kind (key in types dict)) """ p = config.config() repo = ctx.repo() def read(f, sections=None, remap=None): if f in ctx: try: data = ctx[f].data() except IOError as err: if err.errno != errno.ENOENT: raise # handle missing subrepo spec files as removed ui.warn(_("warning: subrepo spec file \'%s\' not found\n") % repo.pathto(f)) return p.parse(f, data, sections, remap, read) else: raise error.Abort(_("subrepo spec file \'%s\' not found") % repo.pathto(f)) if '.hgsub' in ctx: read('.hgsub') for path, src in ui.configitems('subpaths'): p.set('subpaths', path, src, ui.configsource('subpaths', path)) rev = {} if '.hgsubstate' in ctx: try: for i, l in enumerate(ctx['.hgsubstate'].data().splitlines()): l = l.lstrip() if not l: continue try: revision, path = l.split(" ", 1) except ValueError: raise error.Abort(_("invalid subrepository revision " "specifier in \'%s\' line %d") % (repo.pathto('.hgsubstate'), (i + 1))) rev[path] = revision except IOError as err: if err.errno != errno.ENOENT: raise def remap(src): for pattern, repl in p.items('subpaths'): # Turn r'C:\foo\bar' into r'C:\\foo\\bar' since re.sub # does a string decode. repl = stringutil.escapestr(repl) # However, we still want to allow back references to go # through unharmed, so we turn r'\\1' into r'\1'. Again, # extra escapes are needed because re.sub string decodes. repl = re.sub(br'\\\\([0-9]+)', br'\\\1', repl) try: src = re.sub(pattern, repl, src, 1) except re.error as e: raise error.Abort(_("bad subrepository pattern in %s: %s") % (p.source('subpaths', pattern), stringutil.forcebytestr(e))) return src state = {} for path, src in p[''].items(): kind = 'hg' if src.startswith('['): if ']' not in src: raise error.Abort(_('missing ] in subrepository source')) kind, src = src.split(']', 1) kind = kind[1:] src = src.lstrip() # strip any extra whitespace after ']' if not util.url(src).isabs(): parent = _abssource(repo, abort=False) if parent: parent = util.url(parent) parent.path = posixpath.join(parent.path or '', src) parent.path = posixpath.normpath(parent.path) joined = bytes(parent) # Remap the full joined path and use it if it changes, # else remap the original source. remapped = remap(joined) if remapped == joined: src = remap(src) else: src = remapped src = remap(src) state[util.pconvert(path)] = (src.strip(), rev.get(path, ''), kind) return state def writestate(repo, state): """rewrite .hgsubstate in (outer) repo with these subrepo states""" lines = ['%s %s\n' % (state[s][1], s) for s in sorted(state) if state[s][1] != nullstate[1]] repo.wwrite('.hgsubstate', ''.join(lines), '') def submerge(repo, wctx, mctx, actx, overwrite, labels=None): """delegated from merge.applyupdates: merging of .hgsubstate file in working context, merging context and ancestor context""" if mctx == actx: # backwards? actx = wctx.p1() s1 = wctx.substate s2 = mctx.substate sa = actx.substate sm = {} repo.ui.debug("subrepo merge %s %s %s\n" % (wctx, mctx, actx)) def debug(s, msg, r=""): if r: r = "%s:%s:%s" % r repo.ui.debug(" subrepo %s: %s %s\n" % (s, msg, r)) promptssrc = filemerge.partextras(labels) for s, l in sorted(s1.iteritems()): a = sa.get(s, nullstate) ld = l # local state with possible dirty flag for compares if wctx.sub(s).dirty(): ld = (l[0], l[1] + "+") if wctx == actx: # overwrite a = ld prompts = promptssrc.copy() prompts['s'] = s if s in s2: r = s2[s] if ld == r or r == a: # no change or local is newer sm[s] = l continue elif ld == a: # other side changed debug(s, "other changed, get", r) wctx.sub(s).get(r, overwrite) sm[s] = r elif ld[0] != r[0]: # sources differ prompts['lo'] = l[0] prompts['ro'] = r[0] if repo.ui.promptchoice( _(' subrepository sources for %(s)s differ\n' 'use (l)ocal%(l)s source (%(lo)s)' ' or (r)emote%(o)s source (%(ro)s)?' '$$ &Local $$ &Remote') % prompts, 0): debug(s, "prompt changed, get", r) wctx.sub(s).get(r, overwrite) sm[s] = r elif ld[1] == a[1]: # local side is unchanged debug(s, "other side changed, get", r) wctx.sub(s).get(r, overwrite) sm[s] = r else: debug(s, "both sides changed") srepo = wctx.sub(s) prompts['sl'] = srepo.shortid(l[1]) prompts['sr'] = srepo.shortid(r[1]) option = repo.ui.promptchoice( _(' subrepository %(s)s diverged (local revision: %(sl)s, ' 'remote revision: %(sr)s)\n' '(M)erge, keep (l)ocal%(l)s or keep (r)emote%(o)s?' '$$ &Merge $$ &Local $$ &Remote') % prompts, 0) if option == 0: wctx.sub(s).merge(r) sm[s] = l debug(s, "merge with", r) elif option == 1: sm[s] = l debug(s, "keep local subrepo revision", l) else: wctx.sub(s).get(r, overwrite) sm[s] = r debug(s, "get remote subrepo revision", r) elif ld == a: # remote removed, local unchanged debug(s, "remote removed, remove") wctx.sub(s).remove() elif a == nullstate: # not present in remote or ancestor debug(s, "local added, keep") sm[s] = l continue else: if repo.ui.promptchoice( _(' local%(l)s changed subrepository %(s)s' ' which remote%(o)s removed\n' 'use (c)hanged version or (d)elete?' '$$ &Changed $$ &Delete') % prompts, 0): debug(s, "prompt remove") wctx.sub(s).remove() for s, r in sorted(s2.items()): if s in s1: continue elif s not in sa: debug(s, "remote added, get", r) mctx.sub(s).get(r) sm[s] = r elif r != sa[s]: prompts = promptssrc.copy() prompts['s'] = s if repo.ui.promptchoice( _(' remote%(o)s changed subrepository %(s)s' ' which local%(l)s removed\n' 'use (c)hanged version or (d)elete?' '$$ &Changed $$ &Delete') % prompts, 0) == 0: debug(s, "prompt recreate", r) mctx.sub(s).get(r) sm[s] = r # record merged .hgsubstate writestate(repo, sm) return sm def precommit(ui, wctx, status, match, force=False): """Calculate .hgsubstate changes that should be applied before committing Returns (subs, commitsubs, newstate) where - subs: changed subrepos (including dirty ones) - commitsubs: dirty subrepos which the caller needs to commit recursively - newstate: new state dict which the caller must write to .hgsubstate This also updates the given status argument. """ subs = [] commitsubs = set() newstate = wctx.substate.copy() # only manage subrepos and .hgsubstate if .hgsub is present if '.hgsub' in wctx: # we'll decide whether to track this ourselves, thanks for c in status.modified, status.added, status.removed: if '.hgsubstate' in c: c.remove('.hgsubstate') # compare current state to last committed state # build new substate based on last committed state oldstate = wctx.p1().substate for s in sorted(newstate.keys()): if not match(s): # ignore working copy, use old state if present if s in oldstate: newstate[s] = oldstate[s] continue if not force: raise error.Abort( _("commit with new subrepo %s excluded") % s) dirtyreason = wctx.sub(s).dirtyreason(True) if dirtyreason: if not ui.configbool('ui', 'commitsubrepos'): raise error.Abort(dirtyreason, hint=_("use --subrepos for recursive commit")) subs.append(s) commitsubs.add(s) else: bs = wctx.sub(s).basestate() newstate[s] = (newstate[s][0], bs, newstate[s][2]) if oldstate.get(s, (None, None, None))[1] != bs: subs.append(s) # check for removed subrepos for p in wctx.parents(): r = [s for s in p.substate if s not in newstate] subs += [s for s in r if match(s)] if subs: if (not match('.hgsub') and '.hgsub' in (wctx.modified() + wctx.added())): raise error.Abort(_("can't commit subrepos without .hgsub")) status.modified.insert(0, '.hgsubstate') elif '.hgsub' in status.removed: # clean up .hgsubstate when .hgsub is removed if ('.hgsubstate' in wctx and '.hgsubstate' not in (status.modified + status.added + status.removed)): status.removed.insert(0, '.hgsubstate') return subs, commitsubs, newstate def reporelpath(repo): """return path to this (sub)repo as seen from outermost repo""" parent = repo while util.safehasattr(parent, '_subparent'): parent = parent._subparent return repo.root[len(pathutil.normasprefix(parent.root)):] def subrelpath(sub): """return path to this subrepo as seen from outermost repo""" return sub._relpath def _abssource(repo, push=False, abort=True): """return pull/push path of repo - either based on parent repo .hgsub info or on the top repo config. Abort or return None if no source found.""" if util.safehasattr(repo, '_subparent'): source = util.url(repo._subsource) if source.isabs(): return bytes(source) source.path = posixpath.normpath(source.path) parent = _abssource(repo._subparent, push, abort=False) if parent: parent = util.url(util.pconvert(parent)) parent.path = posixpath.join(parent.path or '', source.path) parent.path = posixpath.normpath(parent.path) return bytes(parent) else: # recursion reached top repo path = None if util.safehasattr(repo, '_subtoppath'): path = repo._subtoppath elif push and repo.ui.config('paths', 'default-push'): path = repo.ui.config('paths', 'default-push') elif repo.ui.config('paths', 'default'): path = repo.ui.config('paths', 'default') elif repo.shared(): # chop off the .hg component to get the default path form. This has # already run through vfsmod.vfs(..., realpath=True), so it doesn't # have problems with 'C:' return os.path.dirname(repo.sharedpath) if path: # issue5770: 'C:\' and 'C:' are not equivalent paths. The former is # as expected: an absolute path to the root of the C: drive. The # latter is a relative path, and works like so: # # C:\>cd C:\some\path # C:\>D: # D:\>python -c "import os; print os.path.abspath('C:')" # C:\some\path # # D:\>python -c "import os; print os.path.abspath('C:relative')" # C:\some\path\relative if util.hasdriveletter(path): if len(path) == 2 or path[2:3] not in br'\/': path = os.path.abspath(path) return path if abort: raise error.Abort(_("default path for subrepository not found")) def newcommitphase(ui, ctx): commitphase = phases.newcommitphase(ui) substate = getattr(ctx, "substate", None) if not substate: return commitphase check = ui.config('phases', 'checksubrepos') if check not in ('ignore', 'follow', 'abort'): raise error.Abort(_('invalid phases.checksubrepos configuration: %s') % (check)) if check == 'ignore': return commitphase maxphase = phases.public maxsub = None for s in sorted(substate): sub = ctx.sub(s) subphase = sub.phase(substate[s][1]) if maxphase < subphase: maxphase = subphase maxsub = s if commitphase < maxphase: if check == 'abort': raise error.Abort(_("can't commit in %s phase" " conflicting %s from subrepository %s") % (phases.phasenames[commitphase], phases.phasenames[maxphase], maxsub)) ui.warn(_("warning: changes are committed in" " %s phase from subrepository %s\n") % (phases.phasenames[maxphase], maxsub)) return maxphase return commitphase