Mercurial > hg > mercurial-source
view mercurial/discovery.py @ 13704:a464763e99f1
dirstate: avoid a race with multiple commits in the same process
(issue2264, issue2516)
The race happens when two commits in a row change the same file
without changing its size, *if* those two commits happen in the same
second in the same process while holding the same repo lock. For
example:
commit 1:
M a
M b
commit 2: # same process, same second, same repo lock
M b # modify b without changing its size
M c
This first manifested in transplant, which is the most common way to
do multiple commits in the same process. But it can manifest in any
script or extension that does multiple commits under the same repo
lock. (Thus, the test script tests both transplant and a custom script.)
The problem was that dirstate.status() failed to notice the change to
b when localrepo is about to do the second commit, meaning that change
gets left in the working directory. In the context of transplant, that
means either a crash ("RuntimeError: nothing committed after
transplant") or a silently inaccurate transplant, depending on whether
any other files were modified by the second transplanted changeset.
The fix is to make status() work a little harder when we have
previously marked files as clean (state 'normal') in the same process.
Specifically, dirstate.normal() adds files to self._lastnormal, and
other state-changing methods remove them. Then dirstate.status() puts
any files in self._lastnormal into state 'lookup', which will make
localrepository.status() read file contents to see if it has really
changed. So we pay a small performance penalty for the second (and
subsequent) commits in the same process, without affecting the common
case. Anything that does lots of status updates and checks in the
same process could suffer a performance hit.
Incidentally, there is a simpler fix: call dirstate.normallookup() on
every file updated by commit() at the end of the commit. The trouble
with that solution is that it imposes a performance penalty on the
common case: it means the next status-dependent hg command after every
"hg commit" will be a little bit slower. The patch here is more
complex, but only affects performance for the uncommon case.
author | Greg Ward <greg@gerg.ca> |
---|---|
date | Sun, 20 Mar 2011 17:41:09 -0400 |
parents | 91cb08a9e7fb |
children | 7abab875e647 |
line wrap: on
line source
# discovery.py - protocol changeset discovery functions # # Copyright 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 node import nullid, short from i18n import _ import util, error def findcommonincoming(repo, remote, heads=None, force=False): """Return a tuple (common, missing roots, heads) used to identify missing nodes from remote. If a list of heads is specified, return only nodes which are heads or ancestors of these heads. """ m = repo.changelog.nodemap search = [] fetch = set() seen = set() seenbranch = set() base = set() if not heads: heads = remote.heads() if repo.changelog.tip() == nullid: base.add(nullid) if heads != [nullid]: return [nullid], [nullid], list(heads) return [nullid], [], [] # assume we're closer to the tip than the root # and start by examining the heads repo.ui.status(_("searching for changes\n")) unknown = [] for h in heads: if h not in m: unknown.append(h) else: base.add(h) heads = unknown if not unknown: return list(base), [], [] req = set(unknown) reqcnt = 0 # search through remote branches # a 'branch' here is a linear segment of history, with four parts: # head, root, first parent, second parent # (a branch always has two parents (or none) by definition) unknown = remote.branches(unknown) while unknown: r = [] while unknown: n = unknown.pop(0) if n[0] in seen: continue repo.ui.debug("examining %s:%s\n" % (short(n[0]), short(n[1]))) if n[0] == nullid: # found the end of the branch pass elif n in seenbranch: repo.ui.debug("branch already found\n") continue elif n[1] and n[1] in m: # do we know the base? repo.ui.debug("found incomplete branch %s:%s\n" % (short(n[0]), short(n[1]))) search.append(n[0:2]) # schedule branch range for scanning seenbranch.add(n) else: if n[1] not in seen and n[1] not in fetch: if n[2] in m and n[3] in m: repo.ui.debug("found new changeset %s\n" % short(n[1])) fetch.add(n[1]) # earliest unknown for p in n[2:4]: if p in m: base.add(p) # latest known for p in n[2:4]: if p not in req and p not in m: r.append(p) req.add(p) seen.add(n[0]) if r: reqcnt += 1 repo.ui.progress(_('searching'), reqcnt, unit=_('queries')) repo.ui.debug("request %d: %s\n" % (reqcnt, " ".join(map(short, r)))) for p in xrange(0, len(r), 10): for b in remote.branches(r[p:p + 10]): repo.ui.debug("received %s:%s\n" % (short(b[0]), short(b[1]))) unknown.append(b) # do binary search on the branches we found while search: newsearch = [] reqcnt += 1 repo.ui.progress(_('searching'), reqcnt, unit=_('queries')) for n, l in zip(search, remote.between(search)): l.append(n[1]) p = n[0] f = 1 for i in l: repo.ui.debug("narrowing %d:%d %s\n" % (f, len(l), short(i))) if i in m: if f <= 2: repo.ui.debug("found new branch changeset %s\n" % short(p)) fetch.add(p) base.add(i) else: repo.ui.debug("narrowed branch search to %s:%s\n" % (short(p), short(i))) newsearch.append((p, i)) break p, f = i, f * 2 search = newsearch # sanity check our fetch list for f in fetch: if f in m: raise error.RepoError(_("already have changeset ") + short(f[:4])) base = list(base) if base == [nullid]: if force: repo.ui.warn(_("warning: repository is unrelated\n")) else: raise util.Abort(_("repository is unrelated")) repo.ui.debug("found new changesets starting at " + " ".join([short(f) for f in fetch]) + "\n") repo.ui.progress(_('searching'), None) repo.ui.debug("%d total queries\n" % reqcnt) return base, list(fetch), heads def findoutgoing(repo, remote, base=None, remoteheads=None, force=False): """Return list of nodes that are roots of subsets not in remote If base dict is specified, assume that these nodes and their parents exist on the remote side. If remotehead is specified, assume it is the list of the heads from the remote repository. """ if base is None: base = findcommonincoming(repo, remote, heads=remoteheads, force=force)[0] else: base = list(base) repo.ui.debug("common changesets up to " + " ".join(map(short, base)) + "\n") remain = set(repo.changelog.nodemap) # prune everything remote has from the tree remain.remove(nullid) remove = base while remove: n = remove.pop(0) if n in remain: remain.remove(n) for p in repo.changelog.parents(n): remove.append(p) # find every node whose parents have been pruned subset = [] # find every remote head that will get new children for n in remain: p1, p2 = repo.changelog.parents(n) if p1 not in remain and p2 not in remain: subset.append(n) return subset def prepush(repo, remote, force, revs, newbranch): '''Analyze the local and remote repositories and determine which changesets need to be pushed to the remote. Return value depends on circumstances: If we are not going to push anything, return a tuple (None, outgoing) where outgoing is 0 if there are no outgoing changesets and 1 if there are, but we refuse to push them (e.g. would create new remote heads). Otherwise, return a tuple (changegroup, remoteheads), where changegroup is a readable file-like object whose read() returns successive changegroup chunks ready to be sent over the wire and remoteheads is the list of remote heads.''' remoteheads = remote.heads() common, inc, rheads = findcommonincoming(repo, remote, heads=remoteheads, force=force) cl = repo.changelog update = findoutgoing(repo, remote, common, remoteheads) outg, bases, heads = cl.nodesbetween(update, revs) if not bases: repo.ui.status(_("no changes found\n")) return None, 1 if not force and remoteheads != [nullid]: if remote.capable('branchmap'): # Check for each named branch if we're creating new remote heads. # To be a remote head after push, node must be either: # - unknown locally # - a local outgoing head descended from update # - a remote head that's known locally and not # ancestral to an outgoing head # 1. Create set of branches involved in the push. branches = set(repo[n].branch() for n in outg) # 2. Check for new branches on the remote. remotemap = remote.branchmap() newbranches = branches - set(remotemap) if newbranches and not newbranch: # new branch requires --new-branch branchnames = ', '.join(sorted(newbranches)) raise util.Abort(_("push creates new remote branches: %s!") % branchnames, hint=_("use 'hg push --new-branch' to create" " new remote branches")) branches.difference_update(newbranches) # 3. Construct the initial oldmap and newmap dicts. # They contain information about the remote heads before and # after the push, respectively. # Heads not found locally are not included in either dict, # since they won't be affected by the push. # unsynced contains all branches with incoming changesets. oldmap = {} newmap = {} unsynced = set() for branch in branches: remotebrheads = remotemap[branch] prunedbrheads = [h for h in remotebrheads if h in cl.nodemap] oldmap[branch] = prunedbrheads newmap[branch] = list(prunedbrheads) if len(remotebrheads) > len(prunedbrheads): unsynced.add(branch) # 4. Update newmap with outgoing changes. # This will possibly add new heads and remove existing ones. ctxgen = (repo[n] for n in outg) repo._updatebranchcache(newmap, ctxgen) else: # 1-4b. old servers: Check for new topological heads. # Construct {old,new}map with branch = None (topological branch). # (code based on _updatebranchcache) oldheads = set(h for h in remoteheads if h in cl.nodemap) newheads = oldheads.union(outg) if len(newheads) > 1: for latest in reversed(outg): if latest not in newheads: continue minhrev = min(cl.rev(h) for h in newheads) reachable = cl.reachable(latest, cl.node(minhrev)) reachable.remove(latest) newheads.difference_update(reachable) branches = set([None]) newmap = {None: newheads} oldmap = {None: oldheads} unsynced = inc and branches or set() # 5. Check for new heads. # If there are more heads after the push than before, a suitable # error message, depending on unsynced status, is displayed. error = None for branch in branches: newhs = set(newmap[branch]) oldhs = set(oldmap[branch]) if len(newhs) > len(oldhs): if error is None: if branch: error = _("push creates new remote heads " "on branch '%s'!") % branch else: error = _("push creates new remote heads!") if branch in unsynced: hint = _("you should pull and merge or " "use push -f to force") else: hint = _("did you forget to merge? " "use push -f to force") if branch: repo.ui.debug("new remote heads on branch '%s'\n" % branch) for h in (newhs - oldhs): repo.ui.debug("new remote head %s\n" % short(h)) if error: raise util.Abort(error, hint=hint) # 6. Check for unsynced changes on involved branches. if unsynced: repo.ui.warn(_("note: unsynced remote changes!\n")) if revs is None: # use the fast path, no race possible on push nodes = repo.changelog.findmissing(common) cg = repo._changegroup(nodes, 'push') else: cg = repo.changegroupsubset(update, revs, 'push') return cg, remoteheads