Mercurial > hg
view mercurial/branchmap.py @ 39627:ee7ee0c516ca
narrow: intersect provided matcher with narrowmatcher in `hg diff`
This provides significant speedups when running diff, and no change in behavior
that I'm aware of (or that the tests found). I tested with a repo that I started
using narrow in after it was created and attempted to run `hg diff -c .` and
similar commands in it on a commit that had files not in the narrowspec.
Timing numbers below, using a similar setup as my previous commits.
before=9db85644, m-u is mozilla-unified at eb39298e432d (flatmanifest) and
0553b7f29eaf (treemanifest). l-d-r is a repo simulating a situation I've
encountered where there's one directory with 30k+ subdirectories. N means
narrow, T means treemanifest. The narrowspec is pretty small when in use, and
importantly the narrowspec is applied *after* doing the initial checkout
(without narrowing), so all of these files exist in the filesystem, which is not
normally the case if someone has been using narrow for the entire life of the
clone.
Anything less than a 5% difference in performance is most likely noise.
diff --git:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 1.292 s +- 0.009 s | 1.295 s +- 0.010 s | 100.2%
m-u | | x | 1.296 s +- 0.042 s | 1.299 s +- 0.026 s | 100.2%
m-u | x | | 1.292 s +- 0.010 s | 1.297 s +- 0.021 s | 100.4%
m-u | x | x | 84.2 ms +- 1.2 ms | 83.6 ms +- 0.2 ms | 99.3%
l-d-r | | | 188.7 ms +- 2.7 ms | 188.8 ms +- 2.0 ms | 100.1%
l-d-r | | x | 189.9 ms +- 1.5 ms | 189.4 ms +- 1.2 ms | 99.7%
l-d-r | x | | 97.1 ms +- 1.0 ms | 87.1 ms +- 1.0 ms | 89.7% <--
l-d-r | x | x | 96.9 ms +- 0.8 ms | 87.2 ms +- 0.7 ms | 90.0% <--
diff -c . --git:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 231.6 ms +- 3.1 ms | 228.9 ms +- 1.6 ms | 98.8%
m-u | | x | 150.5 ms +- 1.7 ms | 150.7 ms +- 1.4 ms | 100.1%
m-u | x | | 233.7 ms +- 2.4 ms | 232.2 ms +- 1.9 ms | 99.4%
m-u | x | x | 126.1 ms +- 1.2 ms | 126.8 ms +- 1.2 ms | 100.6%
l-d-r | | | 82.1 ms +- 2.0 ms | 81.8 ms +- 1.4 ms | 99.6%
l-d-r | | x | 3.732 s +- 0.020 s | 3.746 s +- 0.027 s | 100.4%
l-d-r | x | | 83.1 ms +- 0.8 ms | 107.6 ms +- 2.4 ms | 129.5% <--
l-d-r | x | x | 758.2 ms +- 38.8 ms | 188.5 ms +- 1.8 ms | 24.9% <--
rebase -r . --keep -d .^^:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 5.532 s +- 0.087 s | 5.496 s +- 0.016 s | 99.3%
m-u | | x | 5.554 s +- 0.061 s | 5.532 s +- 0.013 s | 99.6%
m-u | x | | 5.602 s +- 0.134 s | 5.508 s +- 0.035 s | 98.3%
m-u | x | x | 582.2 ms +- 15.2 ms | 572.9 ms +- 12.0 ms | 98.4%
l-d-r | | | 629.5 ms +- 12.3 ms | 622.5 ms +- 7.3 ms | 98.9%
l-d-r | | x | 6.173 s +- 0.062 s | 6.185 s +- 0.076 s | 100.2%
l-d-r | x | | 274.5 ms +- 10.0 ms | 272.1 ms +- 6.2 ms | 99.1%
l-d-r | x | x | 4.835 s +- 0.056 s | 4.826 s +- 0.034 s | 99.8%
status --change . --copies:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 214.4 ms +- 1.4 ms | 212.2 ms +- 1.7 ms | 99.0%
m-u | | x | 130.9 ms +- 1.2 ms | 131.7 ms +- 1.1 ms | 100.6%
m-u | x | | 215.0 ms +- 2.1 ms | 214.9 ms +- 2.7 ms | 100.0%
m-u | x | x | 109.5 ms +- 2.3 ms | 107.8 ms +- 0.9 ms | 98.4%
l-d-r | | | 79.6 ms +- 0.9 ms | 79.8 ms +- 1.6 ms | 100.3%
l-d-r | | x | 3.799 s +- 0.037 s | 3.928 s +- 0.021 s | 103.4% <--?
l-d-r | x | | 82.7 ms +- 0.7 ms | 83.2 ms +- 1.0 ms | 100.6%
l-d-r | x | x | 746.8 ms +- 6.1 ms | 739.0 ms +- 4.2 ms | 99.0%
status --copies:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 1.884 s +- 0.012 s | 1.885 s +- 0.013 s | 100.1%
m-u | | x | 1.897 s +- 0.027 s | 1.909 s +- 0.077 s | 100.6%
m-u | x | | 1.886 s +- 0.021 s | 1.891 s +- 0.030 s | 100.3%
m-u | x | x | 92.0 ms +- 0.7 ms | 92.4 ms +- 0.4 ms | 100.4%
l-d-r | | | 570.3 ms +- 18.7 ms | 552.2 ms +- 4.5 ms | 96.8%
l-d-r | | x | 568.9 ms +- 16.1 ms | 567.2 ms +- 11.9 ms | 99.7%
l-d-r | x | | 171.1 ms +- 2.5 ms | 170.4 ms +- 1.2 ms | 99.6%
l-d-r | x | x | 171.6 ms +- 3.4 ms | 171.5 ms +- 1.7 ms | 99.9%
update $rev^; ~/src/hg/hg{hg}/hg update $rev:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 3.107 s +- 0.017 s | 3.116 s +- 0.012 s | 100.3%
m-u | | x | 2.943 s +- 0.010 s | 2.945 s +- 0.019 s | 100.1%
m-u | x | | 3.116 s +- 0.033 s | 3.118 s +- 0.027 s | 100.1%
m-u | x | x | 318.5 ms +- 2.7 ms | 320.8 ms +- 4.8 ms | 100.7%
l-d-r | | | 428.9 ms +- 4.4 ms | 429.5 ms +- 4.0 ms | 100.1%
l-d-r | | x | 9.593 s +- 0.081 s | 9.869 s +- 0.043 s | 102.9%
l-d-r | x | | 253.2 ms +- 3.6 ms | 254.0 ms +- 2.8 ms | 100.3%
l-d-r | x | x | 1.613 s +- 0.009 s | 1.630 s +- 0.017 s | 101.1%
Differential Revision: https://phab.mercurial-scm.org/D4587
author | spectral <spectral@google.com> |
---|---|
date | Thu, 13 Sep 2018 22:48:27 -0700 |
parents | 8547c8590ac1 |
children | d99a588d8515 |
line wrap: on
line source
# branchmap.py - logic to computes, maintain and stores branchmap for local repo # # Copyright 2005-2007 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 struct from .node import ( bin, hex, nullid, nullrev, ) from . import ( encoding, error, pycompat, scmutil, util, ) from .utils import ( stringutil, ) calcsize = struct.calcsize pack_into = struct.pack_into unpack_from = struct.unpack_from def _filename(repo): """name of a branchcache file for a given repo or repoview""" filename = "branch2" if repo.filtername: filename = '%s-%s' % (filename, repo.filtername) return filename def read(repo): f = None try: f = repo.cachevfs(_filename(repo)) lineiter = iter(f) cachekey = next(lineiter).rstrip('\n').split(" ", 2) last, lrev = cachekey[:2] last, lrev = bin(last), int(lrev) filteredhash = None if len(cachekey) > 2: filteredhash = bin(cachekey[2]) partial = branchcache(tipnode=last, tiprev=lrev, filteredhash=filteredhash) if not partial.validfor(repo): # invalidate the cache raise ValueError(r'tip differs') cl = repo.changelog for l in lineiter: l = l.rstrip('\n') if not l: continue node, state, label = l.split(" ", 2) if state not in 'oc': raise ValueError(r'invalid branch state') label = encoding.tolocal(label.strip()) node = bin(node) if not cl.hasnode(node): raise ValueError( r'node %s does not exist' % pycompat.sysstr(hex(node))) partial.setdefault(label, []).append(node) if state == 'c': partial._closednodes.add(node) except (IOError, OSError): return None except Exception as inst: if repo.ui.debugflag: msg = 'invalid branchheads cache' if repo.filtername is not None: msg += ' (%s)' % repo.filtername msg += ': %s\n' repo.ui.debug(msg % pycompat.bytestr(inst)) partial = None finally: if f: f.close() return partial ### Nearest subset relation # Nearest subset of filter X is a filter Y so that: # * Y is included in X, # * X - Y is as small as possible. # This create and ordering used for branchmap purpose. # the ordering may be partial subsettable = {None: 'visible', 'visible-hidden': 'visible', 'visible': 'served', 'served': 'immutable', 'immutable': 'base'} def updatecache(repo): cl = repo.changelog filtername = repo.filtername partial = repo._branchcaches.get(filtername) revs = [] if partial is None or not partial.validfor(repo): partial = read(repo) if partial is None: subsetname = subsettable.get(filtername) if subsetname is None: partial = branchcache() else: subset = repo.filtered(subsetname) partial = subset.branchmap().copy() extrarevs = subset.changelog.filteredrevs - cl.filteredrevs revs.extend(r for r in extrarevs if r <= partial.tiprev) revs.extend(cl.revs(start=partial.tiprev + 1)) if revs: partial.update(repo, revs) partial.write(repo) assert partial.validfor(repo), filtername repo._branchcaches[repo.filtername] = partial def replacecache(repo, bm): """Replace the branchmap cache for a repo with a branch mapping. This is likely only called during clone with a branch map from a remote. """ rbheads = [] closed = [] for bheads in bm.itervalues(): rbheads.extend(bheads) for h in bheads: r = repo.changelog.rev(h) b, c = repo.changelog.branchinfo(r) if c: closed.append(h) if rbheads: rtiprev = max((int(repo.changelog.rev(node)) for node in rbheads)) cache = branchcache(bm, repo[rtiprev].node(), rtiprev, closednodes=closed) # Try to stick it as low as possible # filter above served are unlikely to be fetch from a clone for candidate in ('base', 'immutable', 'served'): rview = repo.filtered(candidate) if cache.validfor(rview): repo._branchcaches[candidate] = cache cache.write(rview) break class branchcache(dict): """A dict like object that hold branches heads cache. This cache is used to avoid costly computations to determine all the branch heads of a repo. The cache is serialized on disk in the following format: <tip hex node> <tip rev number> [optional filtered repo hex hash] <branch head hex node> <open/closed state> <branch name> <branch head hex node> <open/closed state> <branch name> ... The first line is used to check if the cache is still valid. If the branch cache is for a filtered repo view, an optional third hash is included that hashes the hashes of all filtered revisions. The open/closed state is represented by a single letter 'o' or 'c'. This field can be used to avoid changelog reads when determining if a branch head closes a branch or not. """ def __init__(self, entries=(), tipnode=nullid, tiprev=nullrev, filteredhash=None, closednodes=None): super(branchcache, self).__init__(entries) self.tipnode = tipnode self.tiprev = tiprev self.filteredhash = filteredhash # closednodes is a set of nodes that close their branch. If the branch # cache has been updated, it may contain nodes that are no longer # heads. if closednodes is None: self._closednodes = set() else: self._closednodes = closednodes def validfor(self, repo): """Is the cache content valid regarding a repo - False when cached tipnode is unknown or if we detect a strip. - True when cache is up to date or a subset of current repo.""" try: return ((self.tipnode == repo.changelog.node(self.tiprev)) and (self.filteredhash == \ scmutil.filteredhash(repo, self.tiprev))) except IndexError: return False def _branchtip(self, heads): '''Return tuple with last open head in heads and false, otherwise return last closed head and true.''' tip = heads[-1] closed = True for h in reversed(heads): if h not in self._closednodes: tip = h closed = False break return tip, closed def branchtip(self, branch): '''Return the tipmost open head on branch head, otherwise return the tipmost closed head on branch. Raise KeyError for unknown branch.''' return self._branchtip(self[branch])[0] def iteropen(self, nodes): return (n for n in nodes if n not in self._closednodes) def branchheads(self, branch, closed=False): heads = self[branch] if not closed: heads = list(self.iteropen(heads)) return heads def iterbranches(self): for bn, heads in self.iteritems(): yield (bn, heads) + self._branchtip(heads) def copy(self): """return an deep copy of the branchcache object""" return branchcache(self, self.tipnode, self.tiprev, self.filteredhash, self._closednodes) def write(self, repo): try: f = repo.cachevfs(_filename(repo), "w", atomictemp=True) cachekey = [hex(self.tipnode), '%d' % self.tiprev] if self.filteredhash is not None: cachekey.append(hex(self.filteredhash)) f.write(" ".join(cachekey) + '\n') nodecount = 0 for label, nodes in sorted(self.iteritems()): for node in nodes: nodecount += 1 if node in self._closednodes: state = 'c' else: state = 'o' f.write("%s %s %s\n" % (hex(node), state, encoding.fromlocal(label))) f.close() repo.ui.log('branchcache', 'wrote %s branch cache with %d labels and %d nodes\n', repo.filtername, len(self), nodecount) except (IOError, OSError, error.Abort) as inst: # Abort may be raised by read only opener, so log and continue repo.ui.debug("couldn't write branch cache: %s\n" % stringutil.forcebytestr(inst)) def update(self, repo, revgen): """Given a branchhead cache, self, that may have extra nodes or be missing heads, and a generator of nodes that are strictly a superset of heads missing, this function updates self to be correct. """ starttime = util.timer() cl = repo.changelog # collect new branch entries newbranches = {} getbranchinfo = repo.revbranchcache().branchinfo for r in revgen: branch, closesbranch = getbranchinfo(r) newbranches.setdefault(branch, []).append(r) if closesbranch: self._closednodes.add(cl.node(r)) # fetch current topological heads to speed up filtering topoheads = set(cl.headrevs()) # if older branchheads are reachable from new ones, they aren't # really branchheads. Note checking parents is insufficient: # 1 (branch a) -> 2 (branch b) -> 3 (branch a) for branch, newheadrevs in newbranches.iteritems(): bheads = self.setdefault(branch, []) bheadset = set(cl.rev(node) for node in bheads) # This have been tested True on all internal usage of this function. # run it again in case of doubt # assert not (set(bheadrevs) & set(newheadrevs)) newheadrevs.sort() bheadset.update(newheadrevs) # This prunes out two kinds of heads - heads that are superseded by # a head in newheadrevs, and newheadrevs that are not heads because # an existing head is their descendant. uncertain = bheadset - topoheads if uncertain: floorrev = min(uncertain) ancestors = set(cl.ancestors(newheadrevs, floorrev)) bheadset -= ancestors bheadrevs = sorted(bheadset) self[branch] = [cl.node(rev) for rev in bheadrevs] tiprev = bheadrevs[-1] if tiprev > self.tiprev: self.tipnode = cl.node(tiprev) self.tiprev = tiprev if not self.validfor(repo): # cache key are not valid anymore self.tipnode = nullid self.tiprev = nullrev for heads in self.values(): tiprev = max(cl.rev(node) for node in heads) if tiprev > self.tiprev: self.tipnode = cl.node(tiprev) self.tiprev = tiprev self.filteredhash = scmutil.filteredhash(repo, self.tiprev) duration = util.timer() - starttime repo.ui.log('branchcache', 'updated %s branch cache in %.4f seconds\n', repo.filtername, duration) # Revision branch info cache _rbcversion = '-v1' _rbcnames = 'rbc-names' + _rbcversion _rbcrevs = 'rbc-revs' + _rbcversion # [4 byte hash prefix][4 byte branch name number with sign bit indicating open] _rbcrecfmt = '>4sI' _rbcrecsize = calcsize(_rbcrecfmt) _rbcnodelen = 4 _rbcbranchidxmask = 0x7fffffff _rbccloseflag = 0x80000000 class revbranchcache(object): """Persistent cache, mapping from revision number to branch name and close. This is a low level cache, independent of filtering. Branch names are stored in rbc-names in internal encoding separated by 0. rbc-names is append-only, and each branch name is only stored once and will thus have a unique index. The branch info for each revision is stored in rbc-revs as constant size records. The whole file is read into memory, but it is only 'parsed' on demand. The file is usually append-only but will be truncated if repo modification is detected. The record for each revision contains the first 4 bytes of the corresponding node hash, and the record is only used if it still matches. Even a completely trashed rbc-revs fill thus still give the right result while converging towards full recovery ... assuming no incorrectly matching node hashes. The record also contains 4 bytes where 31 bits contains the index of the branch and the last bit indicate that it is a branch close commit. The usage pattern for rbc-revs is thus somewhat similar to 00changelog.i and will grow with it but be 1/8th of its size. """ def __init__(self, repo, readonly=True): assert repo.filtername is None self._repo = repo self._names = [] # branch names in local encoding with static index self._rbcrevs = bytearray() self._rbcsnameslen = 0 # length of names read at _rbcsnameslen try: bndata = repo.cachevfs.read(_rbcnames) self._rbcsnameslen = len(bndata) # for verification before writing if bndata: self._names = [encoding.tolocal(bn) for bn in bndata.split('\0')] except (IOError, OSError): if readonly: # don't try to use cache - fall back to the slow path self.branchinfo = self._branchinfo if self._names: try: data = repo.cachevfs.read(_rbcrevs) self._rbcrevs[:] = data except (IOError, OSError) as inst: repo.ui.debug("couldn't read revision branch cache: %s\n" % stringutil.forcebytestr(inst)) # remember number of good records on disk self._rbcrevslen = min(len(self._rbcrevs) // _rbcrecsize, len(repo.changelog)) if self._rbcrevslen == 0: self._names = [] self._rbcnamescount = len(self._names) # number of names read at # _rbcsnameslen self._namesreverse = dict((b, r) for r, b in enumerate(self._names)) def _clear(self): self._rbcsnameslen = 0 del self._names[:] self._rbcnamescount = 0 self._namesreverse.clear() self._rbcrevslen = len(self._repo.changelog) self._rbcrevs = bytearray(self._rbcrevslen * _rbcrecsize) def branchinfo(self, rev): """Return branch name and close flag for rev, using and updating persistent cache.""" changelog = self._repo.changelog rbcrevidx = rev * _rbcrecsize # avoid negative index, changelog.read(nullrev) is fast without cache if rev == nullrev: return changelog.branchinfo(rev) # if requested rev isn't allocated, grow and cache the rev info if len(self._rbcrevs) < rbcrevidx + _rbcrecsize: return self._branchinfo(rev) # fast path: extract data from cache, use it if node is matching reponode = changelog.node(rev)[:_rbcnodelen] cachenode, branchidx = unpack_from( _rbcrecfmt, util.buffer(self._rbcrevs), rbcrevidx) close = bool(branchidx & _rbccloseflag) if close: branchidx &= _rbcbranchidxmask if cachenode == '\0\0\0\0': pass elif cachenode == reponode: try: return self._names[branchidx], close except IndexError: # recover from invalid reference to unknown branch self._repo.ui.debug("referenced branch names not found" " - rebuilding revision branch cache from scratch\n") self._clear() else: # rev/node map has changed, invalidate the cache from here up self._repo.ui.debug("history modification detected - truncating " "revision branch cache to revision %d\n" % rev) truncate = rbcrevidx + _rbcrecsize del self._rbcrevs[truncate:] self._rbcrevslen = min(self._rbcrevslen, truncate) # fall back to slow path and make sure it will be written to disk return self._branchinfo(rev) def _branchinfo(self, rev): """Retrieve branch info from changelog and update _rbcrevs""" changelog = self._repo.changelog b, close = changelog.branchinfo(rev) if b in self._namesreverse: branchidx = self._namesreverse[b] else: branchidx = len(self._names) self._names.append(b) self._namesreverse[b] = branchidx reponode = changelog.node(rev) if close: branchidx |= _rbccloseflag self._setcachedata(rev, reponode, branchidx) return b, close def setdata(self, branch, rev, node, close): """add new data information to the cache""" if branch in self._namesreverse: branchidx = self._namesreverse[branch] else: branchidx = len(self._names) self._names.append(branch) self._namesreverse[branch] = branchidx if close: branchidx |= _rbccloseflag self._setcachedata(rev, node, branchidx) # If no cache data were readable (non exists, bad permission, etc) # the cache was bypassing itself by setting: # # self.branchinfo = self._branchinfo # # Since we now have data in the cache, we need to drop this bypassing. if 'branchinfo' in vars(self): del self.branchinfo def _setcachedata(self, rev, node, branchidx): """Writes the node's branch data to the in-memory cache data.""" if rev == nullrev: return rbcrevidx = rev * _rbcrecsize if len(self._rbcrevs) < rbcrevidx + _rbcrecsize: self._rbcrevs.extend('\0' * (len(self._repo.changelog) * _rbcrecsize - len(self._rbcrevs))) pack_into(_rbcrecfmt, self._rbcrevs, rbcrevidx, node, branchidx) self._rbcrevslen = min(self._rbcrevslen, rev) tr = self._repo.currenttransaction() if tr: tr.addfinalize('write-revbranchcache', self.write) def write(self, tr=None): """Save branch cache if it is dirty.""" repo = self._repo wlock = None step = '' try: if self._rbcnamescount < len(self._names): step = ' names' wlock = repo.wlock(wait=False) if self._rbcnamescount != 0: f = repo.cachevfs.open(_rbcnames, 'ab') if f.tell() == self._rbcsnameslen: f.write('\0') else: f.close() repo.ui.debug("%s changed - rewriting it\n" % _rbcnames) self._rbcnamescount = 0 self._rbcrevslen = 0 if self._rbcnamescount == 0: # before rewriting names, make sure references are removed repo.cachevfs.unlinkpath(_rbcrevs, ignoremissing=True) f = repo.cachevfs.open(_rbcnames, 'wb') f.write('\0'.join(encoding.fromlocal(b) for b in self._names[self._rbcnamescount:])) self._rbcsnameslen = f.tell() f.close() self._rbcnamescount = len(self._names) start = self._rbcrevslen * _rbcrecsize if start != len(self._rbcrevs): step = '' if wlock is None: wlock = repo.wlock(wait=False) revs = min(len(repo.changelog), len(self._rbcrevs) // _rbcrecsize) f = repo.cachevfs.open(_rbcrevs, 'ab') if f.tell() != start: repo.ui.debug("truncating cache/%s to %d\n" % (_rbcrevs, start)) f.seek(start) if f.tell() != start: start = 0 f.seek(start) f.truncate() end = revs * _rbcrecsize f.write(self._rbcrevs[start:end]) f.close() self._rbcrevslen = revs except (IOError, OSError, error.Abort, error.LockError) as inst: repo.ui.debug("couldn't write revision branch cache%s: %s\n" % (step, stringutil.forcebytestr(inst))) finally: if wlock is not None: wlock.release()