Mercurial > hg
view mercurial/namespaces.py @ 42285:65b3ef162b39
automation: initial support for running Linux tests
Building on top of our Windows automation support, this commit
implements support for performing automated tasks on remote Linux
machines. Specifically, we implement support for running tests
on ephemeral EC2 instances. This seems to be a worthwhile place
to start, as building packages on Linux is more or less a solved
problem because we already have facilities for building in Docker
containers, which provide "good enough" reproducibility guarantees.
The new `run-tests-linux` command works similarly to
`run-tests-windows`: it ensures an AMI with hg dependencies is
available, provisions a temporary EC2 instance with this AMI, pushes
local changes to that instance via SSH, then invokes `run-tests.py`.
Using this new command, I am able to run the entire test harness
substantially faster then I am on my local machine courtesy of
access to massive core EC2 instances:
wall: 16:20 ./run-tests.py -l (i7-6700K)
wall: 14:00 automation.py run-tests-linux --ec2-instance c5.2xlarge
wall: 8:30 automation.py run-tests-linux --ec2-instance m5.4xlarge
wall: 8:04 automation.py run-tests-linux --ec2-instance c5.4xlarge
wall: 4:30 automation.py run-tests-linux --ec2-instance c5.9xlarge
wall: 3:57 automation.py run-tests-linux --ec2-instance m5.12xlarge
wall: 3:05 automation.py run-tests-linux --ec2-instance m5.24xlarge
wall: 3:02 automation.py run-tests-linux --ec2-instance c5.18xlarge
~3 minute wall time to run pretty much the entire test harness is
not too bad!
The AMIs install multiple versions of Python. And the run-tests-linux
command specifies which one to use:
automation.py run-tests-linux --python system3
automation.py run-tests-linux --python 3.5
automation.py run-tests-linux --python pypy2.7
By default, the system Python 2.7 is used. Using this functionality,
I was able to identity some unexpected test failures on PyPy!
Included in the feature is support for running with alternate
filesystems. You can simply pass --filesystem to the command to
specify the type of filesystem to run tests on. When the ephemeral
instance is started, a new filesystem will be created and tests
will run from it:
wall: 4:30 automation.py run-tests-linux --ec2-instance c5.9xlarge
wall: 4:20 automation.py run-tests-linux --ec2-instance c5d.9xlarge --filesystem xfs
wall: 4:24 automation.py run-tests-linux --ec2-instance c5d.9xlarge --filesystem tmpfs
wall: 4:26 automation.py run-tests-linux --ec2-instance c5d.9xlarge --filesystem ext4
We also support multiple Linux distributions:
$ automation.py run-tests-linux --distro debian9
total time: 298.1s; setup: 60.7s; tests: 237.5s; setup overhead: 20.4%
$ automation.py run-tests-linux --distro ubuntu18.04
total time: 286.1s; setup: 61.3s; tests: 224.7s; setup overhead: 21.4%
$ automation.py run-tests-linux --distro ubuntu18.10
total time: 278.5s; setup: 58.2s; tests: 220.3s; setup overhead: 20.9%
$ automation.py run-tests-linux --distro ubuntu19.04
total time: 265.8s; setup: 42.5s; tests: 223.3s; setup overhead: 16.0%
Debian and Ubuntu are supported because those are what I use and am
most familiar with. It should be easy enough to add support for other
distros.
Unlike the Windows AMIs, Linux EC2 instances bill per second. So
the cost to instantiating an ephemeral instance isn't as severe.
That being said, there is some overhead, as it takes several dozen
seconds for the instance to boot, push local changes, and build
Mercurial. During this time, the instance is largely CPU idle and
wasting money. Even with this inefficiency, running tests is
relatively cheap: $0.15-$0.25 per full test run. A machine running
tests as efficiently as these EC2 instances would cost say $6,000, so
you can run the test harness a >20,000 times for the cost of an
equivalent machine. Running tests in EC2 is almost certainly cheaper
than buying a beefy machine for developers to use :)
# no-check-commit because foo_bar function names
Differential Revision: https://phab.mercurial-scm.org/D6319
author | Gregory Szorc <gregory.szorc@gmail.com> |
---|---|
date | Sat, 27 Apr 2019 11:48:26 -0700 |
parents | 4c0683655599 |
children | 2372284d9457 |
line wrap: on
line source
from __future__ import absolute_import from .i18n import _ from . import ( registrar, templatekw, util, ) def tolist(val): """ a convenience method to return an empty list instead of None """ if val is None: return [] else: return [val] class namespaces(object): """provides an interface to register and operate on multiple namespaces. See the namespace class below for details on the namespace object. """ _names_version = 0 def __init__(self): self._names = util.sortdict() columns = templatekw.getlogcolumns() # we need current mercurial named objects (bookmarks, tags, and # branches) to be initialized somewhere, so that place is here bmknames = lambda repo: repo._bookmarks.keys() bmknamemap = lambda repo, name: tolist(repo._bookmarks.get(name)) bmknodemap = lambda repo, node: repo.nodebookmarks(node) n = namespace("bookmarks", templatename="bookmark", logfmt=columns['bookmark'], listnames=bmknames, namemap=bmknamemap, nodemap=bmknodemap, builtin=True) self.addnamespace(n) tagnames = lambda repo: [t for t, n in repo.tagslist()] tagnamemap = lambda repo, name: tolist(repo._tagscache.tags.get(name)) tagnodemap = lambda repo, node: repo.nodetags(node) n = namespace("tags", templatename="tag", logfmt=columns['tag'], listnames=tagnames, namemap=tagnamemap, nodemap=tagnodemap, deprecated={'tip'}, builtin=True) self.addnamespace(n) bnames = lambda repo: repo.branchmap().keys() bnamemap = lambda repo, name: tolist(repo.branchtip(name, True)) bnodemap = lambda repo, node: [repo[node].branch()] n = namespace("branches", templatename="branch", logfmt=columns['branch'], listnames=bnames, namemap=bnamemap, nodemap=bnodemap, builtin=True) self.addnamespace(n) def __getitem__(self, namespace): """returns the namespace object""" return self._names[namespace] def __iter__(self): return self._names.__iter__() def items(self): return self._names.iteritems() iteritems = items def addnamespace(self, namespace, order=None): """register a namespace namespace: the name to be registered (in plural form) order: optional argument to specify the order of namespaces (e.g. 'branches' should be listed before 'bookmarks') """ if order is not None: self._names.insert(order, namespace.name, namespace) else: self._names[namespace.name] = namespace # we only generate a template keyword if one does not already exist if namespace.name not in templatekw.keywords: templatekeyword = registrar.templatekeyword(templatekw.keywords) @templatekeyword(namespace.name, requires={'repo', 'ctx'}) def generatekw(context, mapping): return templatekw.shownames(context, mapping, namespace.name) def singlenode(self, repo, name): """ Return the 'best' node for the given name. What's best is defined by the namespace's singlenode() function. The first match returned by a namespace in the defined precedence order is used. Raises a KeyError if there is no such node. """ for ns, v in self._names.iteritems(): n = v.singlenode(repo, name) if n: return n raise KeyError(_('no such name: %s') % name) class namespace(object): """provides an interface to a namespace Namespaces are basically generic many-to-many mapping between some (namespaced) names and nodes. The goal here is to control the pollution of jamming things into tags or bookmarks (in extension-land) and to simplify internal bits of mercurial: log output, tab completion, etc. More precisely, we define a mapping of names to nodes, and a mapping from nodes to names. Each mapping returns a list. Furthermore, each name mapping will be passed a name to lookup which might not be in its domain. In this case, each method should return an empty list and not raise an error. This namespace object will define the properties we need: 'name': the namespace (plural form) 'templatename': name to use for templating (usually the singular form of the plural namespace name) 'listnames': list of all names in the namespace (usually the keys of a dictionary) 'namemap': function that takes a name and returns a list of nodes 'nodemap': function that takes a node and returns a list of names 'deprecated': set of names to be masked for ordinary use 'builtin': bool indicating if this namespace is supported by core Mercurial. """ def __init__(self, name, templatename=None, logname=None, colorname=None, logfmt=None, listnames=None, namemap=None, nodemap=None, deprecated=None, builtin=False, singlenode=None): """create a namespace name: the namespace to be registered (in plural form) templatename: the name to use for templating logname: the name to use for log output; if not specified templatename is used colorname: the name to use for colored log output; if not specified logname is used logfmt: the format to use for (i18n-ed) log output; if not specified it is composed from logname listnames: function to list all names namemap: function that inputs a name, output node(s) nodemap: function that inputs a node, output name(s) deprecated: set of names to be masked for ordinary use builtin: whether namespace is implemented by core Mercurial singlenode: function that inputs a name, output best node (or None) """ self.name = name self.templatename = templatename self.logname = logname self.colorname = colorname self.logfmt = logfmt self.listnames = listnames self.namemap = namemap self.nodemap = nodemap if singlenode: self.singlenode = singlenode # if logname is not specified, use the template name as backup if self.logname is None: self.logname = self.templatename # if colorname is not specified, just use the logname as a backup if self.colorname is None: self.colorname = self.logname # if logfmt is not specified, compose it from logname as backup if self.logfmt is None: # i18n: column positioning for "hg log" self.logfmt = ("%s:" % self.logname).ljust(13) + "%s\n" if deprecated is None: self.deprecated = set() else: self.deprecated = deprecated self.builtin = builtin def names(self, repo, node): """method that returns a (sorted) list of names in a namespace that match a given node""" return sorted(self.nodemap(repo, node)) def nodes(self, repo, name): """method that returns a list of nodes in a namespace that match a given name. """ return sorted(self.namemap(repo, name)) def singlenode(self, repo, name): """returns the best node for the given name By default, the best node is the node from nodes() with the highest revision number. It can be overriden by the namespace.""" n = self.namemap(repo, name) if n: # return max revision number if len(n) > 1: cl = repo.changelog maxrev = max(cl.rev(node) for node in n) return cl.node(maxrev) return n[0] return None